Dietary Interventions in Atopic Dermatitis: A Comprehensive Scoping Review and Analysis

Abstract

Background

This scoping review aims to critically assess gaps in the current literature on atopic dermatitis (AD) by evaluating the overall effectiveness of dietary interventions. Through a comprehensive analysis that follows the Preferred Reporting Item for Systematic Review and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines, we conducted a thorough search on the Web of Science database in May 2023 using specific search strategies to identify all relevant studies on the research topic.

Summary

A total of 104 full-text articles were included for review. Our synthesis identified seven notable categories of dietary interventions for AD, showcasing the diversity of interventions utilized. This includes vitamin supplementation, probiotic and prebiotic supplementation, dietary fat, biological compounds, foods from natural sources, major nutrients, and diet-related approaches. Further analyses stratified by targeted populations revealed a predominant focus on pediatrics, particularly in probiotic supplementation, and on adults, with an emphasis on vitamin D and E supplementation.

Key Messages

Despite most dietary interventions demonstrating overall effectiveness in improving AD severity and its subjective symptoms, several significant gaps were identified. There was a scarcity of studies on adults and whole-diet interventions, a prevalence of short-term interventions, heterogeneity in study outcomes, designs, and population, occasional disparity between statistical significance and clinical relevance, and a lack of a comprehensive multidisciplinary approach. Nonetheless, these findings offer valuable insights for future AD research, guiding additional evidence-driven dietary interventions and informing healthcare professionals, researchers, and individuals, advancing both understanding and management of AD.

Introduction

Background

Atopic dermatitis (AD) is a chronic, persistent, and pruritic inflammatory skin condition that represents a substantial burden on global public health [1]. While AD typically manifests during early childhood to affect the pediatric population, it can persist into adulthood or even first present in the later stage of life [2]. The multifaceted nature of AD underscores the complex interplay of genetic predispositions [3, 4], immunological responses [5], environmental influences [6], and even lifestyle factors [7]. Emerging epidemiological findings and randomized controlled trials (RCT) in recent decades strongly suggest that diet may play a pivotal role in not only triggering and exacerbating AD but also in its effective management [8–12]. Therefore, research focusing on dietary interventions in AD is essential for advancing treatment strategies, refining our understanding of how diet impacts symptom control and trigger reduction, and ultimately improving the quality of life for individuals with AD.

Definition of AD

Given the highly heterogeneous and complex nature of AD, establishing a standardized definition of AD is crucial for this review. Within this review, we defined AD to be a composite of characteristic clinical features (pruritus, dry or scaly skin, erythema, specific lesions distributed in the flexural areas such as the bends of elbows and knees, as well as the face and neck), disease course (chronic or relapsing) and associated with a personal and/or family history of atopic conditions (such as allergic rhinitis, asthma, or allergic sensitization to common environmental allergens). This is in accordance with the validated guidelines of the Hanifin and Rajka criteria [13] and the UK Working Party’s diagnostic criteria [14]. Adhering to these established criteria ensures a standardized and consistent approach to defining and differentiating AD from other inflammatory skin diseases within the scope of this review. While our primary interest was in examining the influence of dietary interventions on AD symptoms, we took a broader approach to outcomes. In addition to evaluating changes in AD symptoms, we considered alterations in immune responses and skin parameters of individuals with AD. This expanded scope allows us to comprehensively assess the impact of dietary interventions, not only on the manifestations of AD but also on the underlying immune mechanisms and overall skin health. By analyzing a wider range of outcomes, we aim to provide a more nuanced understanding of the potential multifaceted effects of dietary interventions in the context of AD.

Research Gaps and Aims

In this scoping review, we strategically targeted specific gaps in the existing literature to contribute valuable insights to the field of AD. Our primary aim was to address the gap pertaining to the overall effectiveness of dietary interventions for AD. Through a comprehensive evaluation of the existing literature, we aim to present a narrative synthesis that captures the current and updated dietary interventions. This synthesis seeks to enhance the understanding of how specific dietary interventions impact the onset, exacerbation, and management of AD across diverse and specific populations. Additionally, our focus extends to investigating the gap related to the diversity of food and dietary patterns utilized to improve AD. Recognizing the varied approaches to dietary interventions, our review seeks to identify and analyze the spectrum of nutrients, foods, and dietary patterns implicated in the management of AD. Lastly, our aim was to delineate the characteristics of the studied population by identifying and analyzing their demographics, such as age group and geographical location. This involves a detailed examination of the participant profiles within the included studies, providing insights into the diverse demographic factors influencing the effectiveness and diversity of dietary interventions for AD. Ultimately, this scoping review aims to critically emphasize the need for additional and evidence-driven RCTs that empower healthcare professionals, researchers, and individuals with AD in advancing the understanding and management of AD.

Methodology

Search Strategy

This review was conducted in accordance with the Preferred Reporting Item for Systematic Review and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) 2018 guidelines [15] (refer to online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000535903 for the PRISMA-ScR checklist) to ensure methodological rigor and transparency. We conducted a comprehensive literature search to identify dietary intervention studies relevant to AD. The primary literature search was performed on May 10, 2023, using the Web of Science database. Search results were limited to English journal articles published between 1990 and May 2023. To ensure an exhaustive search capturing all relevant studies on the research topic, we utilized 12 well-defined full search terms related to AD and dietary intervention. The search terms included combinations using appropriate Boolean operators such as (i) ((Atopic dermatitis) AND (diet)), (ii) ((Atopic dermatitis) AND (nutrients)), (iii) ((Atopic dermatitis) AND (eating patterns)), (iv) ((Atopic dermatitis) AND (dietary habits)), (v) ((Atopic dermatitis) AND (food groups)), (vi) ((Atopic dermatitis) AND (supplements)), (vii) ((Atopic dermatitis) AND (micronutrients)), (viii) ((Atopic dermatitis) AND (vitamins)), (ix) ((Atopic dermatitis) AND (minerals)), (x) ((Atopic dermatitis) AND (dietary fat)), (xi) ((Atopic dermatitis) AND (dietary protein)) and (xii) ((Atopic dermatitis) AND (dietary carbohydrate)). The search terms center on dietary aspects, including dietary habits, nutrition, various dietary components, dietary patterns, food groups, and dietary supplements to investigate their impact on AD. The initial search yielded a total of 3,022 articles. Following this, a deduplication process was performed to identify and remove duplicate records. Duplicate articles with the same title and author published in the same journal were identified and consolidated, resulting in the removal of 609 duplicate records. The deduplication process ensured that each unique study was included only once in the subsequent stages of the review.

Eligibility Criteria

In accordance with predetermined eligibility criteria for relevance, two researchers (J.J.L. and M.H.L.) independently conducted a thorough assessment of all retrieved articles. Any discrepancies that arose during this process were resolved through internal discussion until a consensus was reached. For this review, inclusion criteria were defined based on our characterization of AD and the stated aims. The scope encompassed studies adhering to a nutritional intervention trial design, focusing on human populations of all age groups. Specifically, eligible studies were required to address AD as a primary condition and employ relevant and appropriate outcome measures to assess the impact of dietary interventions. Full-text journal articles available in English were included. Exclusion criteria were applied to studies that met at least one of the following: pertained to nonhuman subjects (e.g., in vitro studies, murine experiments, canine/feline AD), focused on food allergies or other irrelevant skin diseases (e.g., psoriasis, acne vulgaris, seborrheic dermatitis), primarily investigated the medicinal treatments for AD, cosmetics or beauty-related interventions, or involved a cross-sectional examination of epidemiological dietary factors. Using these eligibility criteria, we initially screened 2,413 records for suitability, excluding 2,313 irrelevant articles. Based on the title and abstract, an additional 18 articles were excluded from the review as they met at least one of the following criteria: the article was a review, had poorly specified study design and methods, and presented outcomes unrelated to AD. A visual representation of the article selection process is presented in Figure 1 alongside the full search terms and filters applied.

Fig. 1.

PRISMA flowchart illustrating the search strategy from Tricco et al. (2018). PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and explanation. Ann Intern Med. 2018;169(7):467–473. doi:10.7326/M18-085.

Data Extraction and Narrative Synthesis

A standardized, pre-piloted form was utilized to extract data from the included studies for assessment of study quality and evidence synthesis. The following study characteristics were abstracted from the selected fill-text articles which included the author(s), publication year, study title and abstract, participant characteristics (e.g., origin, sample size, targeted group), interventions (e.g., study design, study duration, intervention-of-interest), main outcomes of interest (e.g., subjective symptoms, skin parameter measurements, immune changes), and conclusions. A narrative synthesis approach was employed to summarize and analyze the findings of all included studies.

Results

Literature Search

The searches yielded 3,022 eligible journal articles during the primary search and screening process (Fig. 1). After the removal of 609 duplicates, we carefully reviewed the title and abstract of the remaining articles. Of 2,413 publications, only 100 met the inclusion criteria and were included in the present review. The majority of the excluded studies were primarily reviews, not pertaining to human subjects, and had a primary focus on food allergies as the sole outcome. Therefore, we have narrowed down the pool of potentially eligible full-text articles to 84, aligning with the specific aims of this review. In several studies, the research focus and terminology revolved around “eczema” and “atopic eczema.” A rigorous assessment was conducted to validate that these terminologies used in the studies align with the predefined definition of AD, specifically denoting a recurrent itchy rash predominantly distributed in the flexural regions. Upon further review and additional relevant studies, 104 studies were included in the final review.

Overview of Study Characteristics

The descriptive characteristics of the 104 included studies are summarized in Table 1. Among the 104 included studies, the majority (n = 77) adhered to the gold standard of a randomized, double-blind, placebo-controlled design. This suggests most included studies have adopted a robust approach to minimizing biases and ensuring the reliability of their dietary interventions. Five studies used a randomized, placebo-controlled design but were single-blinded while another four were only randomized and double-blinded. Additionally, eight studies were based on open trials. The remaining ten studies contained a variety of study designs, including a mixture of match-paired, crossover, and pre-post studies without clearly specifying the blinding status. Most studies (n = 81) reported a positive improvement in the outcomes, highlighting the effectiveness of dietary interventions. Only a smaller number of studies (n = 23) indicated no significant changes or improvements in their reported outcomes related to AD severity. Study populations originated from 28 countries with most studies being Eurocentric (n = 53). Among the studies conducted in Europe, a notable proportion was based in specific countries: Finland (n = 11), Italy (n = 10), Germany (n = 7), United Kingdom (n = 6), and Norway (n = 5). On the other hand, there was also a considerable number of studies conducted in Asia (n = 28) with eleven focusing on the Japanese and five focusing on Koreans. Surprisingly, no studies were conducted in Singapore or Malaysia despite the high prevalence rate of AD in these highly urbanized countries [16–18]. Thus, this underscores a potential research gap in certain regions, emphasizing the need for increased research focus and awareness to investigate dietary interventions and AD. The number of studies has increased over time from 1990 to 2023. There were nine studies published during 1990–2000 and substantially increased to forty-seven studies during 2001–2011 to indicate a growing interest in the role of diet in managing AD. More than half of the studies focused on the pediatric population, and pregnant and/or lactating women (n = 73) while fewer studies (n = 31) focused on the AD adult population.

Table 1.

Overview of the articles on atopic dermatitis (AD) and dietary interventions published between 1990 and 2023

Characteristics	Overall articles included (n = 104)
Study design, n (%)
Randomized, double-blind, placebo-controlled	77 (74.0)
Randomized, single-blind, placebo-controlled	5 (4.81)
Randomized, double-blind only	4 (3.85)
Open trials	8 (7.69)
Others	10 (9.62)
Treatment effects, n (%)
Positive improvements in AD severity and/or immune responses	81 (77.9)
No significant changes in any measured outcomes	23 (22.1)
Geographical regions, n (%)
Asia Pacific (Central and South Asia, Northeastern Asia, Southeast Asia)	28 (26.9)
Europe (Northern, Southern, Eastern, Western Europe)	53 (51.0)
Americas (North America, South America, Central America, Caribbean)	8 (7.69)
Oceania	6 (5.77)
Middle East and Africa	8 (7.69)
Not stated	1 (0.96)
Years, n (%)
1990–2000	9 (8.65)
2001–2011	47 (45.2)
2012–20231	48 (46.2)
Targeted population, n (%)
Adult (aged ≥18 years)	31 (29.8)
Pregnant and/or lactating women	20 (19.2)
Pediatrics (aged <18 years)	53 (51.0)

¹Studies included in year 2023 are up till May.

Overview of Diet and Foods

The dietary interventions identified in this review were broadly grouped into seven main categories, reflecting diversity in the types of intervention utilized. These categories included interventions involving vitamin supplementation, probiotic and prebiotic supplementation, dietary fat manipulation, biological compounds, consumption of foods from natural sources, major nutrients, and diet-related approaches (Fig. 2). Particularly, there was a growing interest in the proportion of intervention studies involving probiotics, prebiotics, and vitamins over the years. In the same period spanning from 1990 to 2023, interventions centered around diet-related approaches became obsolete while there were more studies involving biological compounds, major nutrients, and natural sources (Fig. 3). Because of the diversity of dietary interventions, measured outcomes, study design, and intervention duration, a meta-analysis for the majority of the included studies was not feasible. Therefore, we have employed a scoping review approach instead to thoroughly examine the extensive body of research within the field of dietary interventions on AD.

Fig. 2.

Summary of dietary interventions categorized into seven main types which include vitamin supplementation, probiotic and prebiotic supplementation, dietary fat manipulation, utilization of biological compounds, consumption of foods from natural sources, major nutrients, and dietary approaches. *The total studies for both probiotics and prebiotics intervention.

Fig. 3.

Summary of the changing trend for dietary intervention studies on atopic dermatitis (AD) published in three periods (1990–2000, 2001–2011, and 2012–2023 May) based on seven main dietary intervention categories. A dotted line was used to demarcate the various categories of dietary intervention.

AD and Dietary Interventions among Various Targeted Population

Although dietary interventions have a role in managing AD symptoms, the effectiveness may vary among different populations due to differences in age, dietary habits, and immune health. Here, we stratified and analyzed various targeted population groups to assess the specific dietary interventions employed for managing AD.

Adult Population

Thirty-one articles focused on the adult population and these studies examined a variety of dietary interventions [19–49]. Among these interventions, vitamins D and E supplementation were the most frequently studied (n = 8), followed by probiotic and prebiotic (n = 7), dietary fat (n = 6), foods of natural source (n = 4), with limited studies focusing on major nutrients, biological compounds, and diet-related approach (n = 2) (Table 2). Generally, vitamin D3 supplementation resulted in improvement in AD symptoms [20–22, 24] with only one study showing no significant changes to AD severity or antimicrobial peptide expression in AD skin [19]. Two separate studies demonstrated the effectiveness of vitamin E supplementation in improving the extent of eczematous lesions in AD adults [25, 26]. For dietary fats, some studies suggested that n-3 fatty acids supplementation modulates immune responses in adults with AD by potentially regulating the proliferation of CD25⁺ T cells [27] and immunoglobulin E (IgE) production [29]. Supplementation of various labeled strains of Lactobacillus plantarum was a common approach observed in three separate studies in the adult population and demonstrated some effectiveness in reducing AD severity and pro-inflammatory cytokine levels [34–36]. Interestingly, interventions involving foods from natural sources, such as traditional herbal medicine [42], deep-sea water [43], fig leaf tea [44], and dodder seed extract [45], were exclusively examined in the adult population. Although these studies reported discernible improvements in skin symptoms, it is crucial to interpret the findings cautiously due to the limited availability and uniqueness of these foods in a typical diet. A habitual intake of a vegetarian diet was effective in improving AD severity by modulating inflammation through the reduction in the numbers of eosinophils, neutrophils, and serum IgE levels [49].

Table 2.

Summarized characteristics of journal articles on atopic dermatitis (AD) and dietary interventions focusing on adults published between 1990 and 2023

Study design	Main objective	Study population and sample size	Diet factor-of-interest (type, dose)	Duration	Main disease/outcome studied	Main clinical effects (changes reported)		Study reference (year)
Vitamins (n = 8)
Randomized, double-blind, placebo-controlled	To determine if supplementation with oral vitamin D alters AMP production in AD skin	USA	Vitamin D3	1 capsule/day for 3 weeks consecutively	AD skin AMPs and IL-13 expression	No changes in CAMP, and IL-13 expression	CAMP expression	Hata et al. [19] (2014)
							[p values were not shown]
		30 (non-atopic vs. AD subjects)	(4,000 IU/capsule)				IL-13 expression
							[p = 0.06]
	To assess if vitamin D3 improves the response to treatment of patients with moderate-to-severe AD	Mexico	Vitamin D3	1 capsule/day for 3 months consecutively	AD severity	↓ in mean patient SCORAD score	SCORAD	Sánchez-Armendáriz et al. [20] (2014)
		58 (29 treated vs. 29 placebo)	(5,000 IU/capsule)				Before [38.0±4.4] and after [20.0±2.2] with [p < 0.002]
	To evaluate the effect of vitamin D supplementation on AD patients	Iran	Vitamin D3	1 capsule/day for 60 days consecutively	AD severity	↓ in mean patient	SCORAD	Amestejani et al. [21] (2012)
							Before [24.8±4.1] and after [15.3±3.1] with [p = 0.01]
		60 (30 treated vs. 30 placebo)	(1,600 IU/capsule)			SCORAD score and total patient TIS score	TIS
							Before [3.5±0.5] and after [1.9±0.4] with [p = 0.032]
	To assess the effects of vitamins D and E supplementation on the clinical manifestation of AD	Iran	Vitamin D3	1 capsule/day or 2 softgels/day for 60 days consecutively	AD severity	↓ in mean patient SCORAD score	SCORAD	Javanbakht et al. [22] (2011)
							Vitamin D
							Before [36.0±3.7] and after [23.3±2.8]
		52					Vitamin E
			(1,600 IU/capsule)				Before [33.3±3.6] and after [20.4±2.4]
							Vitamin D and E
		(13 treated with both D and E)				↓ in intensity (lichenification and pruritus)	Before [35.6±3.7] and after [12.5±2.3]
							[All with p = 0.01]
			Vitamin E				Intensity
		(13 treated with D only)					Vitamin D
							Before [7.4±0.7] and after 4.5±1.8]
							Vitamin E
		(13 treated with E only)				Significant correlations between SCORAD and intensity	Before [5.7±0.6] and after [4.3±0.5]
			(600 IU for two softgels)				Vitamin D and E
							Before [7.0±0.8] and after [2.6±0.4]
		(13 placebo)					[All with p = 0.04]
							Correlation
							[p < 0.001]
	To determine the effects of vitamins E and/or D on erythrocyte SOD and catalase activities in AD patients	Iran	Vitamin D3	1 capsule/day or 2 softgels/day for 60 days consecutively	SOD and catalase activities	SOD activities ↑ all treated groups	SOD activities	Javanbakht et al. [23] (2010)
							Vitamin D
		45					Before [1,124.7±54.6] and after [1,357.3±32.6] with [p = 0.002]
							Vitamin E
			(1,600 IU/capsule)				Before [1,187.5±82.9] and after [1,458.1±129.5] with [p = 0.016]
		(11 treated with both D and E)				Catalase activities ↑ only in those treated with vitamin D3	Vitamin D and E
							Before [1,072.2±45.6] and after [1,277.8±55.1] with [p = 0.015]
		(12 treated with D only)	Vitamin E				Catalase activities
							Vitamin D
							Before [183.1±14.2] and after [219.5±15.9] with [p = 0.026]
		(11 treated with E only)				Significant correlation between SOD activity and serum 25(OH)D	Vitamin D and E
			(600 IU for two softgels)				Before [217.8±20.4] and after [267.7±22.2] with
		(11 placebo)					[p = 0.004]
							Correlation
							[r = 0.378, p = 0.01]
Not specified	To correlate vitamin D concentrations in patients who had AD and to determine if vitamin D supplementation affects the clinical manifestations of AD	Poland	Vitamin D3	1 capsule/day for 3 months consecutively	AD severity	↓ in mean patient SCORAD score	SCORAD	Samochocki et al. [24] (2013)
							Before [45.12±16.07] and after [25.70±11.10] with [p = 0.001]
		(95 AD subjects vs. 58 healthy controls)	(2,000 IU/capsule)		Total IgE level	↓ in mean patient total IgE level	Total IgE (IU/mL)
							Before [1,147.56±1,883.85] and after [994.86±1,680.71] with [p = 0.001]
Randomized, single-blind, placebo-controlled, pilot study	To compare the effects of placebo and vitamin E intake on symptoms and serum IgE levels of AD patients	Europe	Vitamin E	1 pill/day for 8 months consecutively	Exacerbation of eczematous lesions	↓ eczematous lesions and serum IgE level	Lesions	Tsoureli-Nikita et al. [25] (2002)
							23/50 subjects showed great improvement
		96 (50 treated vs. 46 placebo)	(400 IU/pill)				Serum IgE (IU/mL)
					Serum IgE
							62% decrease from 1,005 to 490
							Note: p values were not provided
Randomized, double-blind, placebo-controlled	To evaluate the effect of oral vitamin E on AD treatment	Isafahan, Iran	Vitamin E	1 pill/day for 4 months consecutively	AD severity	↓ in mean patient SCORAD score	SCORAD	Jaffary et al. [26] (2015)
		70 (35 treated vs. 35 placebo)	(400 IU/pill)				Before [11.12] and after [3.89] with [p < 0.05]
Dietary fats ( n = 6)
Randomized, double-blind	To examine the effect of dietary supplementation of n-3 fatty acids on immune changes	Norway	N-3 fatty acid (1 g of highly concentrated ethyl esters of very long-chain n-3 fatty acids)1	6 capsules/day for 4 months	T cell proliferation	↓ in CD25+ lymphocytes	CD25 + (%)	Soyland et al. [27] (1994)
							Before [40.5] and after [35.5] with [p < 0.05]
							Correlation
		21 (10 treated with n-3 fatty acids vs. 11 corn oil controls)			Cytokine measurement	A significant correlation between diet-induced decrease in CD25+ cells and TNF secretion
							(CD25 + cells/TNF)
							[r = 0.80, p < 0.01]
Randomized, double-blind, multicenter study	To investigate if fish oil/corn oil had a beneficial effect on AD	Norway			Clinical symptoms	↓ in mean erythema, induration, pruritus, lichenification, scaling, and area affected	Erythema	Soyland et al. [28] (1994)
							Before [5.0±0.3] and after [3.6±0.3] with [p < 0.001]
							Induration
							Before [4.0±0.3] and after [2.8±0.3] with [p < 0.05]
							Pruritus
							Before [5.9±0.3] and after [4.0±0.3] with [p < 0.001]
		145 (57 treated with fish oil vs. 63 corn oil controls)					Lichenification
							Before [5.0±0.3] and after [3.2±0.3] with [p < 0.01]
							Scaling
							Before [5.0±0.3] and after [3.0±0.3] with [p < 0.01]
							Area affected
							Before [2.0±0.2] and after [1.2±0.2] with [p < 0.05]
Randomized, double-blind, placebo-controlled	To determine the efficacy of dietary n-3 PUFA DHA in AD patients	Berlin, Germany	Docosahexaenoic acid (DHA)	1 pill/day for 8 weeks consecutively	AD severity	↓ in mean patient SCORAD score and IgE levels	SCORAD	Koch et al. [29] (2008)
							Before [37.0] and after [33.4] with [p = 0.009]
		53 (28 treated vs. 25 placebo)	(5.4 g/pill)		IgE synthesis		IgE (ng/mL)
							Before [44.2±38.3] and after [15.2±12.9] with [p = 0.013]
Randomized, parallel, double-blind, placebo-controlled	To investigate the effects of seed and pulp oils of sea buckthorn on AD	Turku, Finland	Sea buckthorn seed (Hippophaë rhamnoides) and pulp oils	10 capsules/day for 4 months	AD severity	↓ in mean patient SCORAD score with sea buckthorn pulp oil treatment	SCORAD	Yang et al. [30] (1999)
		45 (27 treated vs. 18 placebo)					Before [37.2±17.7] and after [29.2±20.8] with [p < 0.01]
	To investigate whether evening primrose oil was effective in AD	India	Epogam	6 capsules twice/day for 16 weeks	AD severity	No statistically significant difference in skin parameters from placebo	Epogam	Berth-Jones et al. [31] (1993)
		123	(500 mg evening primrose oil)				[p = 0.74]
		(41 treated with Epogam)	Efamol				Efamol
		(41 treated with Efamol)
		(41 given marine placebo)	(430 mg evening primrose oil and 107 mg marine fish oil)				[p = 0.26]
Randomized, double-blind, placebo-controlled	To evaluate the efficacy and safety of evening primrose oil in Korean patients with AD	Korea	Evening primrose oil	8 capsules/day for 4 months consecutively	AD severity	↓ in mean patient EASI score by month 4	EASI	Chung et al. [32] (2018)
							Before [4.69±1.63] and after [2.80±0.87] with [p = 0.01]
					TEWL		TEWL (at forearms)
							Experimental [−3.05±8.26] versus control [−1.60±7.10] with [p = 0.714]
		50 (25 treated vs. 25 placebo)	(450 mg/capsule)		Skin hydration	Improvement in skin parameters is no statistically significant	Skin hydration (at forearms)
							Experimental 6.57±11.94] and control [3.03±9.06] with [p = 0.470]
					Subjective pruritus		Subjective pruritus
							Before [4.3±1.1] and after [4.1±1.4] with [p = 0.343]
Probiotics and prebiotics (n = 7)
Randomized, double-blind, placebo-controlled	To investigate the clinical effect of a supplementary diet containing heat-killed lactic acid bacterium (LAB) on adult patients with AD	Japan	Lactobacillus paracasei K71 (100 mg/dose with ∼2.0 × 1011 CFU)	100 mg probiotic powder and 400 mg dextrin over 12 weeks	AD severity	↓ in mean patient skin severity score and improvement in quality of life	Skin severity scores	Moroi et al. [33] (2011)
							[−27.1% reduction] with [p < 0.05]
		34 (17 treated vs. 17 placebo)			Quality of life		QOL scores
							[−29.3% reduction] with [p < 0.05]
Randomize, double-blind, placebo-controlled	To assess changes in SCORAD and immune responses in adults with mild-to-moderate AD after LP IS-10506 supplementation	Surabaya, Indonesia	Lactobacillus plantarum (LP) (IS-10506)	Daily dose over 8 weeks	AD severity	↓ in mean patient skin severity scores and	SCORAD	Prakoeswa et al. [34] (2022)
							Before [34.79±12.41] and after [9.6133±2.552] with [p < 0.001]
					Cytokines		IL-4 (pg/mL)
		30 (15 treated vs. 15 placebo)	(2.0 × 1010 CFU/dose)				Before [3.32±0.52] and after [0.414±0.2336] with [p < 0.001]
					Expression		IL-17 (pg/mL)
							Before [5.50±2.36] and after [2.236±2.000] with [p < 0.001]
Randomized, double-blind, placebo-controlled	To evaluate the efficacy of a food supplement containing selected strains of probiotics in ameliorating AD symptoms in adults	Italy	Probiotic mixture consisting of	One capsule of probiotic mix daily for 56 days	AD severity	↓ in mean patient SCORAD scores	SCORAD	Michelotti et al. [35] (2021)
			1.0 × 109 CFU				Before [20.9±0.5] and after [13.7±0.6] with [p < 0.001]
							TEWL
			L. plantarum PBS067		Skin measurements		Before [14.6±1.3] and after [12.0±1.0] with [p < 0.001]
			1.0 × 109 CFU				Skin moisture
							Before [23.1±0.9] and after [29.4±1.2] with [p < 0.001]
		80 (40 treated vs. 40 placebo)	L. reuteri PBS072			Improvement in TEWL, skin moisturization, and ↓ in TNF-α, TARC, and TSLP expression	TNF-α (pg/mL)
							Before [108.2±9.3] and after [72.7±6.1] with [p < 0.001]
			1.0 × 109 CFU		Inflammatory markers in AD adults		TARC (pg/mL)
							Before [23.8±0.2] and after [22.3±0.1] with [p < 0.001]
			L. rhamnosus				TSLP (pg/mL)
							Before [28.9±0.5] and after [25.9±0.4] with [p < 0.001]
Randomized, single-blind, placebo-controlled	To determine the effects of probiotics on the clinical symptoms, immune responses, and gut microbiota in AD patients	China	1.0 × 109 CFU	Daily dosage of indicated probiotic lyophilized powder over 8 weeks	AD severity	CCF8610 treatment ↓ in mean patient SCORAD scores and IL-10 expression	SCORAD	Fang et al. [36] (2020)
		109					CCF8610
			Lactobacillus plantarum (CCFM8610)				[p < 0.05]
		(43 treated with CCFM8610)			Quality of life		IL-10 (pg/mL)
							CCF8610
			1.0 × 109 CFU			CCFM8610 and CCFM16 changed gut microbiota composition to increase Bifidobacterium
		29 treated with CCFM16)					[p < 0.001]
					Gut microbiome		Bifidobacteria composition
		(26 treated with placebo)	Bifidobacterium bifidum F35 (CCFM16)				B. bifidum and B. animalis subsp. Lactis increased significantly
		(11 treated with oligose)					[p < 0.05]
Randomized, double-blind, placebo-controlled	To evaluate the clinical and immunological effects of the intake of probiotic strain in the treatment of adult patients with moderate or severe AD	Italy	1.0 × 109 CFU Lactobacillus salivarius	Twice per day for 16 weeks	AD severity	↓ in mean patient SCORAD and DLQI scores in the treatment group at the end of 16 weeks	SCORAD	Drago et al. [37] (2011)
							Before [27.57±3.4] and after [13.14±0.27] with [p < 0.001]
					Quality of life
		38 (19 treated vs. 19 placebo)	(LS01) in 1 capsule			No changes to the serum IgE and IL-4/IFN-γ/IL-5/IL-12	DLQI
					Cytokine expression
							Before [8.28±1.79] and after [4.42±0.27] with [p = 0.04]
Randomized, double-blind, placebo-controlled	To evaluate the clinical efficacy of an intake of a combination of 2 probiotics for the treatment of adult AD patients	Italy	Probiotic mixture consisting of	Each with a dose of 1.0 × 109 CFU/day for 12 weeks	AD severity	↓ in mean patient SCORAD and DLQI scores	SCORAD	Iemoli et al. [38] (2012)
							After 3 months
							Before [46.25±3.70] and after [29.45±2.01] with [p < 0.0001]
			Lactobacillus salivarius				DLQI
							After 3 months
			(LS01 DSM2275)			↓ in microbial translocation, immune activation	Before [9.16±0.80] and after [6.58±1.25] with [p = 0.021]
							Microbial translocation
		46 (31 treated vs. 15 placebo)	Bifidobacterium breve		Quality of life		[p = 0.050]
							Immune activation
						Improved Th17Treg and Th1/Th2 ratios from baseline after 3-month treatment	[p < 0.001]
			(BR03 DSM 11604)				Th17/Treg ratio
							[p = 0.029]
							Th1/Th2 ratio
							[p = 0.028]
Randomized, double-blind, placebo-controlled	To examine the effects of the probiotic Bifidobacterium animalis subsp. lactis LKM512 on adult-type AD	Japan	Bifidobacterium animalis subsp. Lactis	Each with a dose of 6.0 × 109 CFU/day for 8 weeks	Pruritus	↓ in mean patient itch	Itch	Matsumoto et al. [39] (2014)
		44 (22 treated vs. 22 placebo)	(LKM512)				[p < 0.05]
Major nutrients, excluding fats (n = 2)
Randomized, double-blind, placebo-controlled	To examine the effect of CTP on inflammation in AD	Japan	Collagen tripeptide (CTP)	3.9 g/day for 12 weeks	AD severity	↓ in mean patient SCORAD scores, TEWL, expression of selected cytokines, and STAT1 signaling	SCORAD	Hakuta et al. [40] (2017)
							[p = 0.002]
							TEWL
					Skin measurements		[p = 0.049]
							TARC (pg/mL)
							[p < 0.01]
		13 (7 treated vs. 6 placebo)			Cytokine expression		MDC (pg/mL)
							[p < 0.0001]
							TSLP (pg/mL)
					TEWL		[p < 0.05]
							STAT1 signaling
							[p < 0.01]
Randomized, placebo-controlled	To investigate whether sucrose is an aggravating factor in AD	Germany	Sucrose	100 g/day in ice-cold black tea for 10 days	AD severity	No statistically significant changes in mean patient SCORAD scores and ECP levels	SCORAD	Ehlers et al. [41] (2001)
							Before [28.3] and after [28.9] with [p > 0.05]
		20 adults			ECP expression		ECP (μg/L)
							Before [1.17] and after [1.22] with [p > 0.05]
Natural sources (n = 4)
Randomized, double-blind, placebo-controlled	To evaluate the efficacy and safety of Hochu-ekki-to in the management of Kikyo patients with AD	Japan	Traditional herbal medicine	7.5 g extract over 24 weeks	AD severity	↓ in total equivalent amount of topical agents used but no statistically difference in skin severity scores	Topical agent usage	Kobayashi et al. [42] (2008)
		78 (38 treated vs. 40 placebo)	Hochu-ekki-to		Use of topical agents		[p < 0.05]
Not specified	To study if drinking deep-sea water improves mineral imbalance in mild-to-moderate AEDS	Japan	Deep-sea water	500 mL/day for 6 months	Skin symptoms	Improvement in skin symptoms	Skin symptoms	Hataguchi et al. [43] (2005)
							27/33 patients showed improvement in their skin symptoms
		33
							[p < 0.05]
Randomized, double-blind, placebo-controlled	To evaluate the safety and AD-relieving effects of prolonged fig leaf tea consumption in patients with mild AD	Osaka, Japan	Fig leaf tea	500 mL/day over 12 weeks	AD severity	↓ in EASI and POEM scores only	EASI	Abe et al. [44] (2022)
					TARC expression		Before [5.83±3.89] and after [3.63±1.96] with [p < 0.05]
		30 (15 treated vs. 15 placebo)			IgE levels		POEM
					Th1/Th2 ratio		Before [9.8±5.9] and after [7.5±7.1] with [p < 0.05]
Randomized, double-blind, placebo-controlled	To assess the efficacy and safety of whey associated with dodder seed extract in the treatment of moderate-to-severe AD adults	Iran	Dodder seed extract	Daily dosage of 4 × 500 mg over 30 days	Skin measurements	Improvement in skin moisture, elasticity, pruritus, and experienced lesser sleep disturbance	Skin moisture	Mehrbani et al. [45] (2015)
							Before [13.43±2.08] and after [29.91±1.68] with [p < 0.001]
							Skin elasticity
							Before [56.04±5.32] and after [79.01±2.96] with [p < 0.001]
		52 (26 treated vs. 26 placebo)	(Cuscuta campestris Yuncker)				Pruritus
							Before [5.7±0.45] and after [2.04±0.32] with [p < 0.001]
							Sleep disturbance
							Before [2.41±0.57] and after [0.66±0.20] with [p < 0.005]
Biological compounds (n = 2)
Randomized, double-blind, placebo-controlled	To investigate the role of ingested histamine as an aggravating factor in adult patients with AD	Berlin, Germany	Histamine	Maintain histamine-free diet for 2 weeks prior to the challenge of histamine-hydrochloride	AD severity	↓ in SCORAD scores and DAO activity	SCORAD	Worm et al. [46] (2009)
							Before [47.0±6.0] and after [31±4.0] with [p = 0.002]
		36 AD adults			Diamine oxidase activity		DAO activity
							[p = 0.036]
Randomized, double-blind, placebo-controlled, crossover	To examine the effects of daily oral L-histidine supplementation on disease severity in adult AD patients	Manchester, England	L-histidine	1 sachet/day for 4 weeks crossed over to placebo (4 g erythritol) for the next 4 weeks	AD severity	↓ in SCORAD and POEM scores	SCORAD	Tan et al. [47] (2017)
							A 34.0% reduction in SCORAD with [p = 0.0029]
		21 AD adults	(4 g/sachet)				POEM
							A 39.0% reduction in POEM with [p = 0.001]
Diet-related approach (n = 2)
Open trial	To investigate whether dietary supplementation with PUFA of the omega-3 series and omega 6 series, vitamins, and minerals have a clinical effect on AD	Oslo, Norway	Supplements consisting of	Daily supplementation for 16 weeks	AD severity	↓ in mean patient SCORAD scores	SCORAD	Eriksen et al. [48] (2006)
			GLA, LA, ALA, EPA, and DHA
		17 AD adults	Vitamins C and E				Almost 50.0% reduction in SCORAD with [p < 0.001]
			Zinc (23 mg)
	To assess whether a vegetarian diet is effective for AD and identify the immunological mechanisms	Japan	Vegetarian diet consisting of	Vegetarian diet daily for 2 months	AD severity	↓ in mean patient SCORAD scores, eosinophil counts, LDH5 activity, NK activity, peripheral neutrophil counts, PGE2 synthesis, and PGE2/monocyte ratio	SCORAD	Tanaka et al. [49] (2001)
							Before [49.9±18.6] and after [27.4±16.8] with [p < 0.001]
					Immune cell counts		Eosinophil count (cells/μL)
			Breakfast				Before [423±367] and after [213±267] with [p < 0.01]
							LDH5 (IU/mL)
					Cytokine expression		Before [45.8±21.3] and after [36.0±12.8] with [p < 0.01]
			Fresh vegetable juice (correspond to 250 g of fresh vegetables)				Serum IgE (IU/mL)
							Before [6,990±8,362] and after [6,234±7,222] with [p > 0.05]
		20 AD adults					NK activity
			Lunch and dinner Brown rice porridge (correspond to 80 g brown rice) sprinkled with 5 g of kelp powder, 200 g tofu, and 10 g of sesame paste		Serum IgE		Before [46.5±5.3] and after [14.1±8.9] with [p < 0.001]
							Neutrophil count (cells/μL)
							Before [3,893±2,512] and after [2,007±884] with [p < 0.001]
							PGE2 (pg/mL)
					NK cells and LDH5 activities		Before [2,886±1,443] and after [1,390±773] with [p < 0.001]
			A daily requirement of 2.5 g of non-refined salt was added				PGE2/monocyte (pg/mL)
							Before [0.0078±0.0023] and after [0.0041±0.0021] with [p < 0.05]

Targeted group (adults aged ≥18 years, n = 31). AD, atopic dermatitis; ALA, alpha-linolenic acid; AMP, antimicrobial peptide; CAMP, cathelicidin; CD, cluster of differentiation; CFU, colony-forming unit; DAO, diamine oxidase; DHA, docosahexaenoic acid; DLQI, dermatology life quality index; EASI, eczema area and severity index; ECP, eosinophilic cationic protein; EPA, eicosatetraenoic acid; GLA, gamma-linolenic acid; IFN-γ, interferon-gamma; IgE, immunoglobulin E; IL, interleukin; IU, international unit; KYNA, kynurenic acid; LA, linolenic acid; LDH5, lactate dehydrogenase 5; MDC, macrophage-derived chemokine; n-3, Omega-3; NK, natural killer; PGE2, prostaglandin E2; POEM, patient-oriented eczema measurement; PUFA, polyunsaturated fatty acids; QOL, quality of life; SCORAD, SCORing atopic dermatitis; SOD, superoxide dismutase; STAT1, signal transducer and activator of transcription 1; TARC, thymus and activation-regulated chemokine; TEWL, transepidermal water loss; Th, t-helper; TIS, three-item severity score; TNF, tumor necrosis factor; Treg, regulatory T cell; TSLP, thymic stromal lymphopoietin.

Pregnant and/or Lactating Women

Twenty articles focused on pregnant and/or lactating women population and these studies examined fewer dietary interventions [50–69] (Table 3). Among these, probiotic and prebiotic supplementation (n = 12) emerged as the most frequently studied intervention. Specifically, seven studies [56–62] within this group focused on Bifidobacterium spp. which was less examined in the adult population. While vitamin supplementation (n = 2) was less explored with only one study showing a significantly reduced risk of AD in infants observed at the early age of 12 months [50]. Prenatal diets supplemented with dietary fat (n = 4) were observed to have limited impacts on the prevalence of AD in infants until a certain age [52, 53]. Two separate studies investigating maternal diet elimination on common food allergens such as eggs [68], cow’s milk, and fish [69] reported a reduction in the incidence of AD among infants.

Table 3.

Summarized characteristics of journal articles on atopic dermatitis (AD) and dietary interventions focusing on pregnant and lactating women published between 1990 and 2023

Study design	Main objective	Study population and sample size	Diet factor-of-interest (type, dose)	Duration	Main disease/outcome studied	Main clinical effects (changes reported)		Study reference (year)
Vitamins (n = 2)
Randomized, double-blind, placebo-controlled	To examine the influence of maternal cholecalciferol supplementation during pregnancy on the risk of atopic eczema in the offspring at ages 12, 24, and 48 months	United Kingdom	Vitamin D3	1 capsule/day from 14 weeks’ gestation till delivery	AE risk	Lower odds ratios of AE only at age 12 months	Risk of AE	El-Heis et al. [50] (2022)
							At 12 months
		703 (352 treated vs. 351 placebo)					[OR: 0.55; 95% CI: 0.32–0.97; p = 0.04]
			(1,000 IU/capsule)
							At 24 months
							[OR: 0.75; 95% CI: 0.37–1.52; p > 0.05]
	To elucidate whether maternal vitamin D supplementation during lactation improves infantile eczema and subsequent allergic disorders	Japan	Vitamin D3	1 capsule for 6 weeks consecutively	Infantile eczema severity	No statistically significant differences between SCORAD scores of vitamin D and placebo group	SCORAD	Norizoe et al. [51] (2014)
		164 (82 treated vs. 82 placebo)	(800 IU/capsule)				Vitamin D3 treated (4.75±7.49) versus placebo (3.95±5.36) with [p > 0.05]
Dietary fats (n = 4)
Randomized, double-blind, placebo-controlled	To investigate the effects of the LC-PUFA supplementation on IgE-associated diseases last up to 2 years of age	Sweden	n-3 fatty acids (EPA, DHA)	9 capsules/day for 5.3 weeks (SD 3.4)	IgE-associated diseases	Lowered cumulative incidence of IgE-associated diseases	Incidence of IgE-associated disease	Furuhjelm et al. [52] (2010)
							6/54 (13.0%)
							[p = 0.01]
		120 (54 treated vs. 66 placebo)			Severity of infant allergic disease	Higher proportions of DHA and EPA in maternal and infant plasma phospholipids were associated with a ↓ severity of allergic phenotypes	Incidence of IgE-associated eczema
							5/54 (9.26%)
							[p = 0.04]
Randomized, parallel, double-blind, placebo-controlled	To assess the effect of dietary supplementation with BCSO on the prevalence of atopy at 12 months of age	Finland	Blackcurrant seed oil (BCSO)	3 g/day during 8th to 16th week of pregnancy	AD prevalence and severity	Lowered prevalence of infant AD at 12 months	Prevalence of AD	Linnamaa et al. [53] (2010)
							33/100 children with AD in BCSO group
							[p = 0.035]
		313 (151 treated vs. 162 placebo)				↓ in mean SCORAD score at 12 months	SCORAD
							80/100 children have a SCORAD class of 1
							[p = 0.035]
	To determine the impact of PUFAs on AD by dietary supplementation of infants	Germany	γ-linolenic acid (GLA)	40 mg GLA/day over first 5 months of pregnancy	AD prevalence	Lowered serum IgE in breastfed infants with AD at age of 12 months	Total serum IgE (IU/mL)	Kitz et al. [54] (2006)
		131 (55 treated vs. 76 placebo)			Total serum IgE		GLA+ (10.0) versus GLA- (125.0) at 12 months
							[p < 0.01]
Randomized, double-blind, placebo-controlled	To assess whether an increased intake of oily fish in pregnancy modifies neonatal immune responses and early markers of atopy	United Kingdom	Salmon fish (150 g/portion)	2 portions/day from 20th week of pregnancy till delivery	Selected cytokine production by CBMC	↓ in IL-2 in response to dermatophagoides pteronyssinus allergen 1 only	IL-2 response	Noakes et al. [55] (2012)
					Total IgE		[p ≤ 0.03]
		123 (62 treated vs. 61 placebo)			AD severity
Probiotics and prebiotics (n = 12)
Open trial	To investigate the effects of bifidobacterial supplementation on the risk of developing allergic diseases in the Japanese population	Wakayama, Japan	5 × 109 CFU	1 g sachet containing 5 × 109 CFU over 4 weeks prior to delivery and 6 months postnatally to their infants	Risk of AD development	↓ in the risk of AD development during the first 18 months of life	Risk of AD	Enomoto et al. [56] (2014)
			Bifidobacterial				At 18 months
							[OR: 0.304; 95% CI: 0.105–0.892; p = 0.033]
		166 (130 treated vs. 36 placebo)			Fecal microbiota composition	↓ in proteobacteria composition among treated mothers	Proteobacteria composition
			Supplementation (B. longum)				At 10 months
							Probiotic group (1.59) versus placebo group (2.77) [p = 0.007]
Randomized, double-blind, placebo-controlled	To investigate whether supplementation of probiotics prevents the development of eczema in infants at high risk	Seoul, South Korea	Probiotic mixture consisting of	Taken daily from 8 weeks before the expected delivery to 3 months after delivery	AD prevalence	↓ in prevalence and cumulative incidence of AD during the first 12 months of life	Prevalence of AD	Kim et al. [57] (2010)
							Probiotic group (18.2%) versus placebo group (40.0%) [p = 0.048]
			1.6 × 109 CFU Bifidobacterium lactis (AD011)
		112 (57 treated vs. 55 placebo)					Cumulative incidence of AD during first 12 months
			1.6 × 109 CFU Lactobacillus acidophilus (AD031)
							Probiotic group (36.4%) versus placebo group (62.9%) [p = 0.029]
Randomized, double-blind, placebo-controlled, parallel	To study the effect of a mixture of 4 probiotic bacterial strains along with prebiotic galacto-oligosaccharides in preventing allergic diseases	Helsinki, Finland	Probiotic capsule consisting of	2 capsules/daily during 2–4 weeks before delivery (mother)	Cumulative incidence of allergic diseases	↓ in prevalence and cumulative incidence of IgE-associated diseases such as eczema and AE	Risk of eczema	Kukkonen et al. [58] (2006)
							[OR: 0.74; 95% CI: 0.55–0.98; p = 0.035]
							Risk of AE
			5.0 × 109 CFU Lactobacillus rhamnosus GG				[OR: 0.66; 95% CI: 0.46–0.95; p = 0.025]
							Prevalence of probiotic bacteria at 6 months
		1,223 (610 treated vs. 613 placebo)			Gut microbiota composition	Lactobacilli and bifidobacterial more frequently colonized the guts of supplemented infants	L. rhamnous GG
			5.0 × 109 CFU Lactobacillus rhamnosus LC705				[RR: 3.96; 95% CI: 2.39–6.55; p < 0.001]
							L. rhamnous LC705
				1 capsule/day mixed with 0.8 g galacto-oligosaccharides for 6 months (infant)
							[RR: 28.3; 95% CI: 3.98–200; p < 0.001]
			2.0 × 108 CFU Bifidobacterium breve Bb99				Propionibacterium JS
							[RR: 14.4; 95% CI: 3.61–57.3; p < 0.001]
							Bifidobacterial total
			2.0 × 109 CFU Propionibacterium freudenreichii ssp. Shermanii JS				[RR: 1.13; 95% CI: 1.01–1.27; p = 0.039]
		Helsinki, Finland			Cumulative incidence of allergic diseases and IgE sensitization	↓ in prevalence of IgE-associated allergic disease in cesarean-delivered children only	Incidence of IgE-associated disease	Kuitunen et al. [59] (2009)
							Probiotic [24.3%] vs. placebo [40.5%]
		891 (445 treated vs. 446 placebo)
							[OR: 0.47; 95% CI: 0.23–0.96; p = 0.035]
Randomized, double-blind, placebo-controlled, parallel	To evaluate a multi-strain, high-dose probiotic in the prevention of eczema	Swansea, Wales	Probiotic capsule consisting of	1 capsule/daily from 36 weeks’ gestation till delivery (mother)	Eczema at age 2 years	No statistically significant between the treated and placebo arms for the cumulative frequency of diagnosed eczema at 2 years	Cumulative frequency of eczema	Allen et al. [60] (2014)
			6.25 × 109 CFU Lactobacillus salivarius (CUL61)				Probiotic [34.1%] versus placebo [32.4%]
		454 (220 treated vs. 234 placebo)	1.25 × 109 CFU	1 capsule/daily from birth to age 6 months (infant)	Incidence of atopic eczema in early childhood		[OR: 1.07; 95% CI: 0.72–1.60]
			Bifidobacterium animalis subsp lactis (CUL34)				Incidence of atopic eczema
			1.25 × 109 CFU				Probiotic [5.3%] vs. placebo [12.1%]
			Bifidobacterium bifidum (CUL20)				[OR: 0.40; 95% CI: 0.18–0.91]
Randomized, double-blind, placebo-controlled	To determine whether probiotic supplementation in early life could prevent development of eczema and atopy at 2 years	New Zealand	6.0 × 109 CFU	Either 1 capsule of HN001 or HN019 daily at 35 weeks’ gestation till delivery (mother)	Cumulative prevalence of eczema	Treated infants had a significantly reduced risk of eczema with HN001 only	Risk of eczema	Wickens et al. [61] (2008)
			Lactobacillus rhamnosus
		512	(HN001)				HN001
		(170 treated with HN001)	9.0 × 109 CFU				[HR: 0.51; 95% CI: 0.30–0.85]
		(171 treated with HN019)	Bifidobacterium animalis subsp. lactis				HN019
		(171 placebo)	(HN019)				[HR: 0.90; 95% CI: 0.58–1.41]
Randomized, double-blind, placebo-controlled	To examine whether Th cell proportions were affected by maternal probiotic supplementation and thus could mediate the preventive effect of probiotics on AD	Trondheim, Norway	Probiotic capsule consisting of	One probiotic capsule per day from 36 weeks’ gestation to 3 months postnatally while breastfeeding	T-helper cell proportions	↓ proportion of Th22 cells in probiotic group	Proportion of Th22 cells	Rø et al. [62] (2017)
			5.0 × 1010 CFU
			Lactobacillus rhamnosus (LGG)
			5.0 × 1010 CFU				Probiotic group [median 0.038%] versus placebo group [median 0.064%]
		415 (211 treated vs. 204 placebo)				Proportion of Th22 was increased in children who developed AD. [p < 0.001]
			Bifidobacterium animalis subsp. lactis (Bb-12)
			5.0 × 109 CFU				[p = 0.009]
			Lactobacillus acidophilus (La-5)
Randomized, double-blind, placebo-controlled, parallel	To prevent eczema and sensitization in infants with a family history of allergic disease by oral supplementation with the probiotic Lactobacillus reuteri	Sweden	1.0 × 108 CFU	5 oil droplets (1 × 108 CFU)/day 4 weeks before birth (mother)	Allergic disease prevalence	Treated group has less IgE-associated eczema during the second year	Incidence of IgE-associated eczema	Abrahamsson et al. [63] (2007)
							Before [20.0%] versus after [8.0%]
		232 (117 treated vs. 115 placebo)	Lactobacillus reuteri	Same amount from birth till 12 months of age (infant)
							[p = 0.02]
Randomized, double-blind, placebo-controlled	To assess the effect on atopic disease of Lactobacillus GG	Turku, Finland	1.0 × 1010 CFU	1.0 × 1010 CFU Lactobacillus GG/day for 2–4 weeks before delivery and postnatally for 6 months	Atopic eczema prevalence	↓ frequency of atopic eczema in the treated group	Incidence of IgE-associated eczema	Kalliomäki et al. [64] (2001)
		159 (77 treated vs. 82 placebo)	Lactobacillus GG				Probiotic group [23.0%] versus placebo group [46%]
Randomized, double-blind, placebo-controlled	To examine whether prenatal treatment with the probiotic Lactobacillus GG can influence the risk of eczema during pregnancy	Melbourne, Australia	1.8 × 1010 CFU	1.8 × 1010 CFU/day from 36 weeks’ gestation until delivery	Risk of eczema during infancy	No association between reduced risk of eczema or IgE-associated eczema and probiotic treatment	Risk of eczema	Boyle et al. [65] (2011)
							Probiotic group [34.0%] versus placebo group [39.0%]
							[RR: 0.88; 95% CI: 0.63–1.22]
		250 (125 treated vs. 125 placebo)	Lactobacillus GG				Risk of IgE-associated eczema
							Probiotic group [18.0%] versus placebo group [19.0%]
							[RR: 0.94; 95% CI: 0.53–1.68]
Randomized, double-blind, placebo-controlled	To evaluate the effectiveness of prenatal and postnatal probiotics in the prevention of early childhood and maternal allergic diseases	Taiwan	1.0 × 1010 CFU	1.0 × 1010 CFU/day from 24 weeks’ gestation until delivery	Cumulative prevalence of sensitization and developing of allergic diseases	No statistically significant effects on sensitization and development of allergic diseases between placebo and LGG group	Incidence of sensitization	Ou et al. [66] (2012)
							At 36 months
							Probiotic group [41.5%] versus placebo group [53.6%]
							[p = 0.27]
		191 (95 treated vs. 96 placebo)	Lactobacillus GG (ATCC 53103)				Incidence of allergic disease
							At 36 months
							Probiotic group [26.8%] versus placebo group [29.3%]
							[p = 0.81]
Randomized, double-blind, placebo-controlled	To evaluate the effect of probiotic Lactobacillus reuteri supplementation on the immunological composition of breast milk in relation to sensitization and eczema in the babies	Sweden	Lactobacillus reuteri (55730)	From 36 weeks’ gestation until delivery (dose not specified)	Immunological composition of breast milk in relation to eczema in babies	↓ TFG-β2 levels and development of sensitization at 24 months	TFG-β2 levels in the colostrum	Böttcher et al. [67] (2008)
							Probiotic group [674] versus placebo group [965]
		109 (54 treated vs. 55 placebo)					[p = 0.02]
							Development of sensitization
							[Adjusted OR: 0.3; 95% CI: 0.1–0.9; p = 0.04]
Diet-related approaches (n = 2)
Randomized, single-blind, parallel	To determine whether egg avoidance diet by nursing mothers should avoid allergenic foods as a preventive sensitization	Indonesia	Eggs (diet elimination)	Avoiding eggs in diets from delivery till 4 months of lactation	AD incidence	↓ the incidence of AD in infants among egg avoidance nursing mothers	AD incidence	Nurani et al. [68] (2008)
		79 (39 avoidance vs. 40 non-avoidance controls)					Non-egg avoidance
							[OR: 14.76; 95% CI: 1.76–123.39; p = 0.010]
Match pair	To determine if a diet free of eggs, cow’s milk, and fish during early lactation reduced atopy/allergic manifestations	Sweden	Eggs, cow’s milk, and fish (diet elimination)	Diet avoidance during the first 3 months of lactation	Allergic manifestations	↓ the incidence of AD in the maternal diet group during the first 6 months postpartum only	Cumulative incidence of AD	Sigurs et al. [69] (1992)
							Intervene (10.8%) versus control (28.0%)
		115 (65 avoidance vs. 50 non-avoidance controls)					[p < 0.05]

Targeted group (pregnant and lactating women, n = 20). AE, atopic eczema; AS, allergic asthma; AR, allergic rhinitis; CBMC, cord blood mononuclear cells; CFU, colony-forming units; DHA, docosahexaenoic acid; EPA, eicosatetraenoic acid; HR, hazard ratio; IgE, immunoglobulin E; IL, interleukin; IU, international unit; LC-PUFA, long-chain polyunsaturated fatty acids; SCORAD, SCORing atopic dermatitis; SD, standard deviation; TFG-β2, transforming growth factor-beta 2, Th, t-helper.

Pediatric Population

The majority of intervention studies focused on the younger population (age <18 years), with a total of fifty-four articles directed toward understanding and managing AD within the pediatric population [70–122] (Table 4). There were a considerable number of studies investigating probiotic and prebiotic supplementation in pediatrics (n = 30). This highlights a strong interest in using gut health and microbial modulation as early-life interventions for AD in the field. Almost all studies focused on two particular genera (Lactobacillus and Bifidobacterium). Of these, sixteen studies explored Lactobacillus rhamnosus (L. rhamnosus) either independently [85–91] or in conjunction with another probiotic strain [100–108]. It is important to note that various studies utilized distinct L. rhamnosus strains such as LGG (ATCC 53103) [85, 86], LCS-742 [102], and 19070-2 [103]. Interestingly, the outcomes revealed a mixed response as certain strains demonstrated improvements in AD severity, while others did not exhibit significant therapeutic effects. This variation underscores the complexity of strain-specific analysis, suggesting that not all strains within the L. rhamnosus species may address AD severity. Three separate studies demonstrated that prebiotic supplementation has effective long-term protective effects on AD and were associated with a reduction in plasma levels of total IgE [112] and a lowered risk of AD [113]. For vitamin supplementation, vitamin D3 (n = 12) [70–80] was exclusively studied in the pediatric population, while investigations on dietary fats (n = 4) explored γ-linolenic acid [81, 82], docosahexaenoic acid [83, 84], and arachidonic acid [83]. Studies on diet-related approaches (n = 5) involved avoidance of monosodium glutamate [118], eggs [120, 122] and/or cow’s milk [120], supplementation with whey or casein [121], and selected foods consumption [119]. These interventions [120–122], however, showed limited improvement in disease severity. A single study employing an open-label, case-control design demonstrated that pancreatic enzyme supplementation represents a safe and effective approach to reducing the severity of AD among affected children [117]. This is the only single study that utilized biological compounds in pediatrics. Conflicting outcomes were observed in two separate studies investigating hydrolyzed protein formulas with one indicating a potential reduction in eczema risk for predisposed children [115] while another highlighted no significant difference in the incidence of AD [116].

Table 4.

Summarized characteristics of journal articles on atopic dermatitis (AD) and dietary interventions focusing on the pediatric population aged <18 years which includes infants, young children, and adolescents, published between 1990 and 2023

Study Design	Main objective		Study population and sample size	Diet factor-of-interest (type, dose)	Duration	Main disease/outcome studied		Main clinical effects (changes reported)		Study reference (year)
Vitamins (n = 11)
Randomized, double-blind, placebo-controlled	To evaluate the impact of vitamin D supplementation on response to standard treatment in pediatrics with severe AD		Egypt	Vitamin D3	2 capsule/day for 12 weeks (with baseline therapy of topical 1% hydrocortisone cream twice daily for 3 months)	AD severity		↓ in the mean EASI scores	EASI	Mansour et al. [70] (2020)
			86 (44 treated vs. 42 placebo)	(1,600 IU/capsule)					Before [44.4±6.28] versus after [20.42±14.6] with [p = 0.035]
Randomized, double-blind, placebo-controlled	To determine the effect of vitamin D supplementation on winter-related AD		Mongolia	Vitamin D3	1,000 IU/day for 1 month	AD severity		↓ in the mean EASI scores	EASI	Camargo et al. [71] (2014)
									Before [21±9] versus after [14.5±8.8] with [p = 0.01]
			107 (58 treated vs. 49 placebo)	(1,000 IU/capsule)		Quality of life		Improvement in IGA at 1 month of treatment	IGA
									[Fisher’s exact test p = 0.03]
									[p for trend = 0.04]
Randomized, double-blind, placebo-controlled	To determine whether vitamin D levels are correlated with AD severity and the effects of vitamin D supplementation on disease modification		Canada	Vitamin D3	2 drops of vitamin D3/day for 3 months	AD severity		No statistically significant difference in the SCORAD scores between supplementation and placebo arm after 3 months of intervention	SCORAD	Lara-Corrales et al. [72] (2019)
			45 (21 treated vs. 24 placebo)	(1,000 IU/drop)					Treatment group [15.35±9.71] versus placebo group [15.13±8.97] with [p = 0.07]
Randomized, double-blind, placebo-controlled	To assess vitamin D3 serum levels in a pediatric population suffering from chronic eczema (IgE-mediated and non-IgE-mediated)		Italy	Vitamin D3	1 capsule/day for 3 months	IgE-associated and non-IgE-mediated chronic eczema and their severity		No statistically significant correlation between vitamin D serum levels, SCORAD index, and total IgE levels in both sensitized and non-sensitized children after 3 months intervention	Correlation	Galli et al. [73] (2015)
									Sensitized eczema children
									Vitamin D (ng/mL)
									[48.0±41.6]
									Total IgE (IU/mL)
									[577.0±994.0]
									SCORAD
									[18.1±17.7]
									[p for comparisons between Vit D and total IgE = 0.41]
			89 (41 treated vs. 48 placebo)	(2,000 IU/capsule)					Non-sensitized eczema children
									Vitamin D (ng/mL)
									[48.8±39.3]
									Total IgE (IU/mL)
									[18.9±9.7]
									SCORAD
									[16.5±16.5]
									[p for comparisons between Vit D and total IgE = 0.15]
Randomized, double-blind, placebo-controlled	To determine the effects of vitamin D supplements on clinical impact including Staphylococcus aureus skin colonization evaluation in AD patients		Thailand	Vitamin D3	1 capsule/day for 4 weeks	AD severity		↓ in SCORAD scores, S. aureus skin colonization and an inverse correlation between S. aureus skin colonization and SCORAD score	SCORAD	Udompataikul et al. [74] (2015)
									[p = 0.028]
									S. aureus colonization
			24 (12 treated vs. 12 placebo)	(2,000 IU/capsule)		S. aureus skin colonization			[p = 0.022]
									Correlation
									[r = −1.0; p < 0.000]
Randomized, double-blind, placebo-controlled	To investigate the relationship between vitamin D3 and AD		India	Vitamin D3	60,000 IU/week for 6 weeks	AD severity		↓ in the mean SCORAD scores	SCORAD	Modi et al. [75] (2021)
									At 4 weeks
			60 (30 treated vs. 30 placebo)						Before [47.8±7.5] versus after [12.8±5.1] with [p = 0.03]
Randomized, double-blind, parallel	To evaluate the effect of symbiotic and vitamin D3 supplements on the severity of AD among infants under 1 year of age		Iran	Vitamin D3 and multi-strain symbiotic mixture consisting of	5 drops/day of symbiotic and 1,000 IU vitamin D3/daily for 2 months	AD severity		↓ in the basal SCORAD scores in both symbiotic and vitamin D3 treatment groups as compared to control group	Symbiotic	Aldaghi et al. [76] (2022)
				2 × 109 CFU of					[p < 0.001]
				Lactobacillus rhamnosus
									Vitamin D3
			81 (27 each for symbiotic, vitamin D3, and control)	Lactobacillus reuteri and Bifidobacterium infantis with the prebiotic fructo-oligosaccharides
									[p = 0.001]
Single-center, prospective, longitudinal	To investigate the correlation between AD and vitamin D deficiency and to examine the possible effect of vitamin D oral supplementation on AD evolution in children through modulation of the immune system		Italy	Vitamin D3	1 capsule/day for 3 months	AD severity		↓ in the mean SCORAD scores	SCORAD	Di Filippo et al. [77] (2015)
									Before [46.13±15.68] versus after [22.57±15.28] with [p < 0.001]
									IL-2 (pg/mL)
									Before [8.22±7.39] versus after [1.24±4.26] with [p < 0.001]
									IL-4 (pg/mL)
			22 AD patients	(1,000 IU/capsule)		Cytokine serum level		↓ in selected cytokine concentration including IL-2, IL-4, IL-6, and IFN-γ	Before [9.01±7.05] versus after [1.36±4.26] with [p < 0.001]
									IL-6 (pg/mL)
									Before [15.11±9.13] versus after [6.81±9.60] with [p = 0.007]
									IFN-γ (pg/mL)
									Before [20.05±22.84] versus after [0.19±0.79] with [p = 0.019]
Open label	To investigate whether oral cholecalciferol supplementation changes stratum corneum expression of VDR, camp/LL-37, and TSLP in children with AD		Chile	Vitamin D3	Age-adjusted weekly oral dose of liquid VD3 for 6 weeks	AD severity		↓ in the mean SCORAD scores	SCORAD	Cabalín et al. [78] (2023)
									Before [41.4±13.5] versus after [31.5±15.8] with [p = 0.0007]
									Camp expression
					8,000 IU/week for 2–5.9 years			↑ epidermal gene expression of camp in both lesional and non-lesional skin	Non-lesional skin
									[p = 0.014]
								LL-37 peptide ↑ only in lesional skin	Lesional skin
			22 (16 moderate AD and vs. 6 severe AD)		12,000 IU/week for 6–11.9 years	Camp/LL-37 and TSLP expression			[p = 0.0007]
									LL-37 peptide
					16,000 IU/week for 12–18 years			TSLP expression did not change significantly	[p = 0.014]
									TSLP expression
									[p > 0.05]
Pre-post intervention	To assess the influence of vitamin D supplementation on the severity of AD		Curitiba, Brazil	Vitamin D	Indicated dosage for 4 weeks followed by 15,000 IU/week maintenance for 2 months subsequently	AD severity		Statistically significant ↓, but not clinically significant in the mean SCORAD scores	SCORAD	Imoto et al. [79] (2021)
				(50,000 IU/week for deficient)
			152	(15,000 IU/week for insufficient)					Before [19.4] versus after [12.3] with [p < 0.01]
Randomized, double-blind, placebo-controlled	To evaluate the clinical effect of oral supplementation of vitamin D3 5,000 IU/day plus basal therapy in an urban population of Mexico City with AD		Mexico	Vitamin D	5,000 IU/day over 3 months	AD severity		↓ in the mean SCORAD scores in patients with AD	SCORAD	Sancgez-Armendariz et al. [80] (2018)
									Sufficient > 30 ng/mL of vitamin D3
			65 (33 treated vs. 32 placebo)						Before [40.4±11.8] versus [19.6±11.6] with [p < 0.001]
Dietary fats (n = 4)
Randomized, double-blind, placebo-controlled, parallel	To study the effect of γ-linolenic acid (GLA) in children with AD and the effect on		Europe	Evening primrose oil	4 capsules twice daily for 16 weeks	Eczema symptoms		No statistically significance difference was found between treatment and placebo groups	Selected clinical assessments, mean differences between groups	Hederos and Berg [81] (1996)
									Redness
									[mean = −1.9]
									Dryness
									[mean = −5.4]
			60 (30 treated vs. 30 placebo)	(500 mg/capsule which contains 400 mg GLA)					Itch
									[mean = −3.0]
									Scaling
									[mean = −1.1]
									[All p > 0.05]
Randomized, parallel, double-blind, placebo-controlled	To investigate the possible preventive effect of GLA supplementation on the development of AD in infants at risk		The Netherlands	GLA	100 mg of GLA/day for the first 6 months of life	Incidence of AD in the first year of life		↓ in the mean SCORAD scores in the GLA supplemented group as compared to the placebo group	SCORAD	Van Gool et al. [82] (2003)
									Treatment group [6.32±5.32] versus placebo group [8.28±6.54]
			121 (61 treated vs. 60 placebo)			AD severity		A significant negative association between increased GLA concentration and AD severity	Association
									[r = −0.233; p = 0.013]
Randomized, double-blind	To investigate the incidence of allergic and respiratory diseases through age 3 years in children fed DHA and ARA supplemented formula during infancy		USA	DHA and ARA	17 mg/100 kcal DHA +34 mg/100 kcal ARA daily for 12 months	Allergic manifestations		Lowered odds for developing wheezing/asthma/AD but not for AD	Odds	Birch et al. [83] (2010)
									Wheezing/asthma/AD
			89 (38 treated vs. 51 placebo)						[OR: 0.25; 95% CI: 0.09–0.67; p = –0.006]
									AD
									[OR: 0.41; 95% CI: 0.14–1.16; p = –0.09]
Controlled, multicenter intervention study	To determine the impact of altered exposure to diet factor during pregnancy and infancy on the incidence of allergy-related diseases at 2 years of age		1,374	Oily fish	5 mL cod liver oil twice/week from 6 months of age for 4–6 weeks	Risk for AD, wheeze, asthma		No statistically significant impact on AD	Incidence of AD	Dotterud et al. [84] (2013)
					(infant)				[Adjusted OR: 0.93; 95% CI: 0.78–1.10]
Probiotics and prebiotics (n = 30)
Open trial	To determine whether oral Lactobacillus rhamnosus GG may act by generating immunosuppressive mediator in atopic children	Turku, Finland		Lactobacillus rhamnosus GG (ATCC 53103)	1.0 × 1010 CFU twice day for 4 weeks	Cytokine production	↓ in the serum IL-10 concentration		IL-10 (pg/mL)		Pessi et al. [85] (2000)
									Before [mean = 10.2] versus after [mean = 4.3] with [p < 0.001]
		9					No statistically significant in fecal sIgA concentration		sIgA (mg/mL)
									Before [mean = 20.4] versus after [mean = 20.9] with [p = 0.89]
Randomized, double-blind, placebo-controlled	To investigate the interaction of Lactobacillus rhamnosus GG with skin and gut microbiota and humoral immunity in infants with AD	Turku, Finland			3.4 × 109 CFU/day over 3 months	Ig secreting cells	↓ in the proportion of IgA-, IgM-secreting cells, and CD19+CD27+ B cells		Baseline-adjusted ratios for treated vs. untreated at 1 month		Nermes et al. [86] (2011)
									IgA-secreting cells
		37 (19 treated vs. 18 placebo)				Bacterial count	No statistically significant differences in the bifidobacterial species composition and mean SCORAD scores		[OR: 0.59; 95% CI: 0.36–0.99; p = 0.044]
						Bacterial count			Proportion of CD19+CD27+ B cells (%)
									Treated [12] versus placebo [7] with [p = 0.009]
Randomized, double-blind, placebo-controlled	To assess the efficacy of oral supplementation of viable and heat-inactivated probiotic bacteria in the management of atopic disease and their effects on gut microbiota composition	Turku, Finland		Lactobacillus rhamnosus GG (LGG)	1.0 × 109 CFU for about 7.5 weeks	AD severity	↓ in the SCORAD scores within the LGG viable group is greater than the placebo group		SCORAD		Kirjavainen et al. [87] (2003)
		35							LGG viable
		(14 treated with viable LGG)							Before [mean = 19] versus after [mean = 5] with [p < 0.05]
									Heat-inactivated
		(13 treated with heat-inactivated LGG)							Before [mean = 15] versus after [mean = 7] with [p < 0.05]
		(8 given placebo)
Randomized, double-blind, placebo-controlled	To reassess the efficacy of orally administered LGG in infants with AD	Germany			5.0 × 109 CFU of LGG twice/day for 8 weeks	AD severity	No statistically significant differences in clinical symptoms, use of topical corticosteroids and antihistamines, and serum IgE level		SCORAD		Fölster-Holst et al. [88] (2006)
						Use of topical corticosteroids (applications per week)			Before [43.3±2.4] versus after [35.3±3.0]
		53 (26 treated vs. 27 placebo)				Use of antihistamines			Topical corticosteroid
									Before [5.6±1.0] versus after [3.0±0.6]
						Serum IgE level			Antihistamines
									Before (10 [39%]) versus after (9 [37%])
									Serum IgE level (IU/mL)
									Treated [341] versus placebo [339]
									[All with p > 0.05]
Randomized, double-blind, placebo-controlled	To investigate the therapeutic effect of LGG as a food supplement in infants suffering from AD	Berlin, Germany			5.0 × 109 CFU of LGG twice/day for 12 weeks	AD severity	No statistically significant improvement in mean SCORAD scores between treated and placebo group		SCORAD		Grüber et al. [89] (2007)
									Treated [19.6±15.4] versus placebo [15.1±12.1]
									Use of medication
						Use of rescue medication (hydrocortisone 1% ointment)			Treated [0.8±45.0] versus placebo [3.5±29.8]
		102 (54 treated vs. 48 placebo)
									Total serum IgE (kU/L )
						Total serum IgE level			Treated [0.17±0.3] versus placebo [0.26±0.45]
									[All with p > 0.05]
Randomized, double-blind, placebo-controlled	To investigate the therapeutic effects of the probiotic LGG in children with AD	Italy			1.0 × 109 CFU of LGG/day for 12 weeks	AD severity (minimal clinically important difference)	Rate of subjects achieving MCID at 16 weeks was higher in the treated group as compared to placebo group		SCORAD		Carucci et al. [123] (2022)
									Percentage of children ≥8.7 units for SCORAD index with [p < 0.05]
						Quality of life			IDQOL
							Improvement in the quality of life for both groups
		100 (50 treated vs. 50 placebo)
						Gut and skin microbiome			Treated; at 16 weeks
									[median = 1; IQR: 3] with [p < 0.05]
Randomized, double-blind, placebo-controlled	To determine if probiotic administration during the first 6 months of life decreases childhood asthma and eczema	Finland			1.0 × 1010 CFU of LGG/day for first 6 months of life	Cumulative incidence of eczema	Probiotic treatment does not prevent the development of eczema at 2 years of age		Incidence of eczema		Cabana et al. [90] (2017)
		184 (92 treated vs. 92 placebo)							[Hazard ratio: 0.95; 95% CI: 0.59–1.53; log-rank p = 0.83]
Randomized, double-blind, placebo-controlled	To determine whether early probiotic supplementation prevents allergic disease in high-risk infants	Perth, Australia		Lactobacillus acidophilus (LAVRI-A1)	3.0 × 109 CFU dissolved in 1–2 mL sterile water/day from birth to 6 months	AD rate, severity	No statistically differences in the rate and severity of AD between treated and placebo groups		Rate		Taylor et al. [91] (2007)
									[p = 0.581]
									Severity
		226 (115 treated vs. 111 placebo)					The proportion of children with AD and sensitization was higher in the probiotic group than the placebo group		[p = 0.995]
									SPT+ AD
									Treated [23/88] versus placebo [12/86] with [p = 0.045]
Randomized, double-blind, placebo-controlled	To examine additional effects of two different dose of paraprobiotic Lactobacillus acidophilus on the clinical treatment in young children afflicted by AD with diagnosed or suspected food allergy	Japan		Lactobacillus acidophilus (L-92)	20 mg (2 × 1010 CFU) of L-92 dry powder/day for 24 weeks	AD severity	↓ in the SCORAD scores, TARC, and total IgE levels		SCORAD		Nakata et al. [92] (2019)
									Before [median: 39.1; range: 12.2–83.8] versus after [8.1±18.5] with [p = 0.040]
									Total IgE (IU/mL)
		59 (29 treated vs. 30 placebo)				Total IgE level			Treated [median: −0.14; range: −0.34 to 0.39] versus placebo [median: −0.01; range: −0.28 to 0.53] with [p = 0.03]
									TARC (pg/mL)
									Treated [median: −504; range: −19,279 to 1,068] versus placebo [median: 86; range: −29,661 to 805] with [p = 0.03]
Randomized, double-blind, placebo-controlled, parallel	To determine if the consumption of heat-inactivated probiotic Lactobacillus paracasei would have a beneficial effect on the clinical symptoms of AD and help limit the quantity of corticosteroids needed as concomitant treatment	Taiwan		Lactobacillus paracasei (GM-080)	1.0 × 1010 CFU/day over 16 weeks	AD severity	↓ in SCORAD and IDQOL scores		SCORAD		Yan et al. [93] (2019)
									[p < 0.001]
									TEWL
						Quality of life			Arms
							Only ↓ in TEWL of the lesional skin on the arms was statistically significant		[p = 0.043]
									Legs
									[p = 0.087]
		126 (64 treated vs. 62 placebo)							Trunk
						TEWL	CCL17 was observed to ↓, but not statistically significance		[p = 0.054]
									IDQOL
									Before [8.59±3.53] versus after [4.95±3.23] with [ p < 0.001]
						CCL17 level			CCL17 (pg/mL)
							Total IgE increased in both treated and placebo group, but not statistically significant		Treated
									Before [186.58±364.84] versus after [133.94±270.42]
						Total IgE			Total IgE (kU/mL)
									Treated
									Before [151.08±260.11] versus after [212.96±365.88]
Randomized, double-blind, placebo-controlled	To characterize the changes caused by Lactobacillus casei	Poland		Lactobacillus casei (DN-114001)	1.0 × 109 CFU/day for 3 months	AD severity	↓ mean SCORAD scores		SCORAD		Klewicka et al. [94] (2011)
	in intestinal microbiota in infants with AD	40 (18 treated vs. 22 placebo)							Before [21.3±9.5] versus after [2.8±3.6] with [p < 0.05]
Randomized, double-blind, placebo-controlled	To determine the underlying immunological effects of probiotics on moderate-to-severe AD	Australia		Lactobacillus fermentum (VR1 003)	1.0 × 109 CFU twice/day for 8 weeks	AD severity	↓ SCORAD scores and was directly correlated with the increase in IFN-γ response to staphylococcus aureus enterotoxin B		SCORAD		Prescott et al. [95] (2005)
									[p < 0.05]
		53 (26 treated vs. 27 placebo)				Cytokine response			Correlation between SCORAD and IFN-γ
									[r = −0.443; p = 0.026]
Randomized, double-blind, placebo-controlled	To evaluate the clinical and anti-inflammatory effect of Lactobacillus pentosus in children with mild-to-moderate AD	South Korea		Lactobacillus pentosus	1.0 × 109 CFU/day for 12 weeks	AD severity	↓ in the mean SCORAD scores		SCORAD		Ahn et al. [96] (2020)
									Before [30.4±8.6] versus after [23.6±11.0] with [p < 0.001]
									TEWL
									Before [30.3±17.0] versus after [28.1±11.3] with [p = 0.444]
						TEWL			Eosinophils (per μL)
									Before [5.7±0.7] versus after [5.8±0.8] with [p = 0.241]
									Total IgE (kU/L)
									Before [4.7±1.5] versus after [4.9±1.5] with [p = 0.122]
									IL-5 (pg/mL)
							No statistically significant differences in the TEWL values, blood eosinophil counts, serum total IgE level, and the selected cytokines
						Blood eosinophil count			Before [3.90±0.00] versus after [4.24±1.19] with [p = 0.075]
									IL-6 (pg/mL)
		82 (41 treated vs. 41 placebo)							Before [0.66±0.92] versus after [0.84±1.65] with [p = 0.740]
									IL-10 (pg/mL)
						Serum total IgE			Before [0.95±0.39] versus after [1.24±1.62] with [p = 0.369]
									IL-13 (pg/mL)
									Before [128.96±24.57] versus after [138.86±71.34] with [p = 0.303]
									IL-17 (pg/mL)
						Cytokine levels			Before [0.43±1.35] versus after [0.49±0.75] with [p = 0.600]
									IFN-γ (pg/mL)
									Before [16.32±4.18] versus after [18.58±12.24] with [p = 0.366]
Randomized, double-blind, placebo-controlled	To assess the probiotic function of Lactobacillus plantarum in children with mild and moderate AD	Indonesia		Lactobacillus plantarum	1.0 × 1010 CFU twice/day for 12 weeks	AD severity	↓ in the mean SCORAD scores, IL-4, IFN-γ, and IL-17 levels		SCORAD		Prakoeswa et al. [97] (2017)
									Treated group [18.533±14.200] versus placebo group [22.040±8.817] with [p = 0.000]
									IL-4 (IU/mL)
									Treated group [4.277±4.892] versus placebo group [5.815±6.633]
						Cytokine levels	No statistically significant changes in serum IgE		IFN-γ (IU/mL)
									Treated group [0.528±0.634] versus placebo group [0.684±1.006] with [p = 0.006]
									IL-17 (IU/mL)
		22 (12 treated vs. 10 placebo)		(IS-10506)					Treated group [0.151±0.135] versus placebo group [0.128±0.134] with [p = 0.000]
									IgE (IU/mL)
						Serum IgE	Higher ratio of Foxp3+/IL-10 in the probiotic group than in placebo group
									Treated group [504.533±415.686] versus placebo group [909.580±885.051] with [p > 0.05]
									Foxp3+:IL-10 (IU/mL)
									Treated group [0.050±0.135] versus placebo group [0.014±0.018] with [p = 0.001]
Randomized, double-blind, placebo-controlled	To evaluate the effect of L. sakei proBio65 live and dead cells when administered for 12 weeks to children and adolescents with AD	South Korea		Lactobacillus sakei	1.0 × 1010 cells/day for 12 weeks	AD severity	↓ in the mean SCORAD and IGA scores for treatments with viable and nonviable strains		SCORAD		Rather et al. [98] (2021)
									Viable group
							Improvement in skin moisture content with the nonviable strain treatment
									[3.08±1.44] with [p = 0.0488]
						Quality of life	Improvement in skin sebum content with both the viable and nonviable strain treatment		Non-viable group [1.48±1.00] with [p = 0.0220]
									IGA
		90 (30 each for live, treated, dead cells, and placebo)							Viable group
									[0.75±0.17] with [p = 0.0005]
						Immune cell count	↓ in the eosinophil count with viable strain treatment		Non-viable group [0.59±0.18] with [p = 0.0039]
									Skin moisture (%)
									[p < 0.0001]
									Skin sebum (%)
									[p < 0.0001]
									Eosinophil count (mm 3 )
									Before [588.75±122.72] versus after [428.75±77.11] with [p = 0.0331]
Randomized, double-blind, placebo-controlled	To investigate whether probiotics could modify the expression of genetic predisposition to eczema conferred by genetic variation in susceptibility genes	Europe		Lactobacillus rhamnosus (HN001)	Either 1 capsule of 6.0 × 109 CFU HN001 or 9.0 × 109 CFU HN019 daily between 2 and 16 days post birth till age 2	Expression of eczema susceptibility SNPs	↓ risk of eczema in the treatment group compared with the placebo group for 16 genetic variants with HN001 and 3 genetic variants with HN019		Eczema		Morgan et al. [99] (2014)
									HN001
									rs7927894 (C11orf30), rs2738182, rs5743399 and rs5743409 (DEFB1), rs1046295 and rs3794379 (PHF11), rs1423001, rs3756688 and rs3815735 (SPINK5), rs1800825 (CCL5), rs2569190 (CD14), rs1800875 (CMA1), rs20541 (IL-13), and rs1801275, rs2057768 and rs210- 7356 (IL4R)
							Furthermore, some genetic variants were associated with a significant ↓ risk of SCORAD ≥10		HN019
				Bifidobacterium animalis subspp lactis (HN019)					rs2738182 and rs5743399 (DEFB1), and rs2107356 (IL4R)
		33							↓ risk of SCORAD ≥ 10
									HN001
									rs5743399 (DEFB1), rs1046295, rs3794378 and rs3794379 (PHF11), rs2287772 and rs4529181 (SPINK5), rs1800825 (CCL5), rs2569190 (CD14), rs180- 0875 (CMA1), rs20541 (IL-13), and rs1801275, rs205- 7768 and rs2107356 (IL4R)
									HN019
									rs5743399 (DEFB1), rs1046295 and rs379- 4378 (PHF11), rs2287772 and rs4529181 (SPINK5), and rs360721 (IL18)
Randomized, double-blind, placebo-controlled	To examine the effect of a combination of two probiotic strains administered in late infancy and early childhood on the development of allergic diseases and sensitization	Denmark		Lactobacillus rhamnosus (LGG)	1.0 × 109 CFU of LGG and BB-12/day over 6 months	Incidence of eczema	↓ incidence of AD		Incidence of AD		Schmidt et al. [100] (2019)
		285 (143 treated vs. 142 placebo)		Bifidobacterium animalis subsp lactis (BB-12)					Treated [4.2%] versus placebo [11.5%] with [p = 0.036]
Randomized, double-blind, placebo-controlled	To evaluate the safety, tolerance, and preventive effect on AD of an experimental α-lactalbumin-enriched and symbiotic-supplemented infant formula	Nantes, France		Lactobacillus rhamnosus	1.4 × 108 CFU of LCS-742 and M63 daily till 6 months of age	Manifestation of AD	↓ frequency of AD at 6 months in the treated group		Manifestation of AD		Rozé et al. [101] (2012)
				(LCS-742)
									Treated [2.60%] versus placebo [17.8%] with [p = 0.03]
				Bifidobacterium longum subsp. infantis (M63)
		97 (48 treated vs. 49 placebo)					A negative correlation between manifestation of AD and colonization of bifidobacterial		Correlation
				96% galacto-oligosaccharides
									[R2 = 0.27; p < 0.005]
				4% short-chain fructo-oligosaccharides
Randomized, double-blind, crossover	To evaluate the clinical and anti-inflammatory effect of probiotic supplementation in children with AD	Denmark		Lactobacillus rhamnosus (19070-2)	Either 1 dose of 1.0 × 109 CFU 19070-2	AD severity	↓ in the extent of eczema, serum eosinophil cationic protein levels		Extent of eczema		Rosenfeldt et al. [102] (2003)
									Before [mean = 18.2%] versus after [mean = 13.7%] with [p = 0.02]
									Serum ECP (μg/L)
						Serum ECP	No statistically significant changes in the cytokine production		Before [52.3±5.5] versus after [34.6±7.6] with [p = 0.03]
									Changes in cytokine
									IL-2 (pg/mL)
									[p = 0.35]
		43		Lactobacillus reuteri (DSM12246)	or DSM12246 twice/day over 20 weeks	Cytokine production	However, there is a significant association between the level of IL-4 and clinical improvement		IL-4 (pg/mL)
									[p = 0.35]
									IL-10 (pg/mL)
									[p = 0.62]
									Association
									[non-parametric correlation factor: −on-parap = 0.028]
Randomized, double-blind, placebo-controlled	To examine the effect of two probiotics (Lactobacillus rhamnosus and Bifidobacteria lactis) on established AD in children	New Zealand		Lactobacillus rhamnosus	1.0 × 1010 CFU/day for 12 weeks	AD severity	↓ geometric mean ratios of SCORAD values among food-sensitized children compared with baseline at the end of 12 weeks treatment		SCORAD		Sistek et al. [103] (2006)
		59 (29 treated vs. 30 placebo)		Bifidobacteria lactis
									Before [baseline: 1.00] versus after [0.73 (0.54–1.00] with [p = 0.047]
Randomized, double-blind, placebo-controlled	To evaluate the clinical and inflammatory effects of supplementation of a hydrolyzed formula with two probiotic strains of bacteria on symptoms of AD in infancy	Netherlands		Lactobacillus rhamnosis (NP-Lrh)	3.0 × 108 CFU/day for 3 months	AD severity	No statistically significant differences in the SCORAD scores, sensitization and eosinophil counts		SCORAD		Brouwer et al. [104] (2006)
									NP-Lrh
									Decrease 0.05/month from baseline of 24.23
		50
									NP-LGG
									Decrease 0.2/month from baseline of 29.4
									Total IgE (kU/L)
		(17 treated with NP-Lrh)		Lactobacillus GG (NP-LGG)		Allergic sensitization			NP-Lrh
									Before [median: 4; range <2–31] versus after [median: 7; range: <2–219]
									NP-LGG
		(16 treated with NP-LGG)							Before [median: 25; range: <2–238] versus after [median: 25; range: <2–408]
									Eosinophils (× 10 9 /L)
									NP-Lrh
		(17 placebo)							Before [median: 0.4; range: 0.1–3.88] versus after [median: 0.26; range: 0.01–1.29]
									NP-LGG
									Before [median: 0.54; range: 0.18–2.76] versus after [median: 0.48; range: 0.26–2.12]
Randomized, double-blind, placebo-controlled	To assess the effect of probiotic supplementation in the first 6 months of life on eczema and allergic sensitization at 1 year of age in Asian infants at risk of allergic disease	Singapore		Lactobacillus rhamnosus (CGMCC1.3724)	Either 1.0 × 107 CFU of BL999 or 2.0 × 107 CFU of CGMCC1.3724 within 12h for the first 6 months of life	Eczema and allergic sensitization	No statistically significant differences in the incidence of eczema, SCORAD scores, and prevalence of allergen sensitization between treated and placebo groups		Incidence of eczema		Soh et al. [105] (2009)
									Treated [27/124; 22%] versus placebo [30/121; 25%]
									Median SCORAD at 12 months
		253 (127 treated vs. 126 placebo)		Bifidobacteria longum (BL999)					Treated [17.10] versus placebo [11.60] with [p = 0.17]
									Rate of sensitization
									Treated [24%] versus placebo [19%]
									[Adjusted OR: 1.43; 95% CI: 0.76–2.70]
Randomized, double-blind, placebo-controlled	To assess the potential of probiotics to control allergic inflammation at an early age	Turku, Finland		Lactobacillus GG	Either 3.0 × 108 CFU of Lactobacillus GG or 1.0 × 109 CFU of Bifidobacterium lactis over 2 months	Extent and severity of eczema	Significant improvement in SCORAD scores		SCORAD		Isolauri et al. [106] (2000)
									LGG treated
									Before [median = 16, IQR = 7–25] versus after [median = 1; IQR = 0.1–8.7] with [p = 0.01]
							Significant ↓ in the urinary eosinophil protein X		Bifidobacterium treated
		27		Bifidobacterium lactis		Concentration of circulating cytokine/chemokine			Before [median = 16, IQR = 7–25] versus after [median = 0; IQR = 0–3.8] with [p = 0.002]
									EPX
							The serum concentrations of IL-1r alpha, TNF-alpha, GM-CSF, sICAM-1, RANTES, and MCP-1alpha were not modified by probiotics		LGG treated
									[p = 0.04]
									Bifidobacterium treated
									[p = 0.01]
Randomized, double-blind, placebo-controlled	To investigate whether probiotic bacteria have any beneficial effect on atopic eczema/dermatitis syndrome	Finland		Lactobacillus rhamnosus (LGG)	Either 1 capsule of 5.0 × 109 CFU of LGG only or a mixture of 5.0 × 109 CFU of LGG with 2 × 108 CFU Bbi99 and 2 × 109 CFU P. JS daily for 4 weeks	AD severity	↓ in the mean SCORAD in IgE-sensitized infants		SCORAD		Viljanen et al. [107] (2005)
									Treated
		230		Bifidobacterium breve (Bbi99)					Before [mean = 37.60 versus after [mean = 11.5]
		(80 treated with LGG alone)							Placebo
		(76 treated with mixed)		Propionibacterium. Freudenreichii					Before [mean = 29.9 versus after [mean = 10.2
									[p = 0.036]
		(74 placebo)		ssp. Shermanii
				JS
Randomized, double-blind, placebo-controlled	To assess the clinical efficacy and impact of 2 probiotic strains with fructo-oligosaccharide on peripheral blood lymphocyte subsets in preschool children with moderate-to-severe AD	Ukraine		Lactobacillus acidophilus (DDS-1)	1 g (5.0 × 109 CFU) twice/day for 8 weeks	AD severity	A significant ↓ in percentages of SCORAD and IDQOL scores		SCORAD		Gerasimov et al. [108] (2010)
									Treated [mean decrease: −14.2; SD: 9.9] versus placebo [mean decrease: −7.8; SD: 7.7] with [p = 0.001]
									IDQOL
						Quality of life			Treated [mean decrease: 33%] versus placebo [mean decrease: 19.0%] with [p = 0.003]
		96 (48 treated vs. 48 placebo		Bifidobacterium lactis (UABLA-12)			A significant correlation between CD4 percentage, CD25 percentage, CD25 absolute count, and SCORAD values		DFI
									Treated [mean decrease: 35.2%] versus placebo [mean decrease: 23.8%] with [p = 0.010]
									Correlation
						Lymphocyte subsets in peripheral blood			CD4% and SCORAD
									[r = 0.642, p < 0.05]
									CD25% and SCORAD
									[r = 0.746, p < 0.05]
									CD25 absolute count and SCORAD
									[r = 0.733, p < 0.05]
Randomized, double-blind, placebo-controlled	To investigate whether dietary supplementation of infants with eczema at age 3–6 months with Lactobacillus paracasei CNCM I-2116 or Bifidobacterium lactis CNCM I-3446 had a treatment effect or altered allergic disease progression	Not stated		Lactobacillus paracasei (CNCM I-2116)	1.0 × 1010 CFU of either strain daily for 3 months	Eczema severity	No statistically significant changes in SCORAD scores, IDQoL scores, percentage of sensitization, and urinary EPX/creatinine concentrations in the treated groups pre- and post-intervention		SCORAD		Gore et al. [109] (2012)
									CNCM 1-2116
									Before [25.4; 95% CI: 22.1–29.0] versus after [12.5; 95% CI: 9.2–16.4]
									CNCM 1-3446
		137				Quality of life			Before [25.9; 95% CI: 22.8–29.2] versus after [4.8; 95% CI: 9.4–16.6]
									IDQOL
									CNCM 1-2116
		(45 treated with CNCM I-2116)		Bifidobacterium lactis (CNCM I-3446)					Before [4.8; 95% CI: 3.8–6.1] versus after [2.9; 95% CI: 2.0–4.0]
									CNCM 1-3446
						Sensitization			Before [5.2; 95% CI: 4.3–6.2] versus after [3.3; 95% CI: 2.4–4.3]
									Sensitization
									CNCM 1-2116
		(45 treated with CNCM I-3446)							Before [67%] versus after [65%]
									CNCM 1-3446
									Before [51%] versus after [62%]
						Urinary EPX/creatinine			Urinary EPX/creatinine (μg/mmol)
		(47 placebo)							CNCM 1-2116
									Before [29.2; 95% CI: 19.1–44.5] versus after [12.7; 95% CI: 7.2–22.4]
									CNCM 1-3446
									Before [18.1; 95% CI: 11.0–30.0] versus after [13.1; 95% CI: 7.7–23.1]
Randomized, double-blind, placebo-controlled, prospective	To investigate the effect of a prebiotic mixture of GOS & FOS on the incidence of AD during the first 6 months of life in formula fed infants at high risk of atopy	Italy		Short-chain galacto-oligosaccharide (GOS)	0.8 g/100 mL GOS and FOS over 6 months	AD development	Lesser infants in the intervention group developed AD at 6 months of age		AD development		Moro et al. [110] (2006)
		259 (129 treated vs. 130 placebo)							Treated [9.8%; 95% CI: 5.4–17.1] versus placebo [23.1%; 95% CI: 16.0–32.1] with [p = 0.014]
Randomized, double-blind, placebo-controlled, prospective	To analyze the effect of GOS and FOS on the immune response in infants with AD at 6 months of age	Italy				Total IgE level	↓ in the plasma level of total IgE, IgG1, IgG2, and IgG3, but not IgG4		IgE (kU/L)		van Hoffen et al. [111] (2009)
									Treated [median: 4.00; IQR: 1.90–8.00] versus placebo [median: 10.0; IQR: 3.0–28.0] with [p < 0.01]
									IgG1 (g/L)
									Treated [median: 2.26; IQR: 1.74–3.34] versus placebo [median: 3.09; IQR: 2.19–4.29] with [p < 0.01]
									IgG2 (g/L)
									Treated [median: 0.66; IQR: 0.51–0.90] versus placebo [median: 0.99; IQR: 0.61–1.26] with [p < 0.01]
		84 (41 treated vs. 43 placebo)		Long-chain fructo-oligosaccharides (FOS)
									IgG3 (g/L)
									Treated [median: 1.04; IQR: 0.64–1.85] versus placebo [median: 1.76; IQR: 1.15–2.77] with [p < 0.01]
									IgG4 (g/L)
									Treated [median: 427.2; IQR: 18.3–1,178.2] versus placebo [median: 187.7; IQR: 5.35–1,455.8] with [p > 0.01]
Randomized, double-blind, placebo-controlled, prospective	To evaluate if the protective effects of GOS and FOS were lasting beyond intervention period	Italy				Cumulative incidence of allergic diseases, including AD	↓ cumulative incidence of AD among the intervention group as compared to those in placebo group		Incidence of AD		Arslanoglu et al. [112] (2008)
									Treated [13.6%] versus placebo [27.9%] with [p < 0.05]
		206 (102 treated vs. 104 placebo)
Randomized, double-blind, placebo-controlled	To evaluate the effect of a postnatal probiotic combination on development of allergic diseases in very preterm infants	Australia		Probiotic mixture consisting of	1.0 × 109 CFU probiotic combination daily from birth to 1 month after delivery	Allergic diseases, including eczema and atopic eczema	No statistically significant difference in eczema, AE, and atopic sensitization incidence between the probiotic and placebo group in the first 2 years of life		Incidence		Plummer et al. [113] (2019)
									Eczema
				Bifidobacterium infantis (BB-02)					Treated [30.0%] versus placebo [27.0%] with [p > 0.05]
									Atopic eczema
		1,099 (548 treated vs. 551 placebo)		Streptococcus thermophilus (TH-4)					Treated [5.0%] versus placebo [2.0%] with [p > 0.05]
									Atopic sensitization
				Bifidobacterium lactis (BB-12)					Treated [13.0%] versus placebo [11.0%] with [p > 0.05]
Major nutrients, excluding fats (n = 2)
Randomized, double-blind	To compare the course of eczema in predisposed children after nutritional intervention to the natural course of eczema		Germany	Hydrolyzed protein formula	Extensively hydrolyzed casein	Physician-diagnosed eczema		Predisposed children without intervention had a higher risk of eczema than those without a familial predisposition	Risk	Berg et al. [114] (2010)
			5,991 (2,252 treated vs. 3,739 non-intervene)						[OR: 2.1; 95% 1.6–2.7]
Randomized, triple-blind, placebo-controlled, prospective	To evaluate the preventive effect of a hydrolyzed protein formula versus an intact protein formula on allergy development in preterm infants with or without risk factors		Italy		Extensively hydrolyzed protein formula for 2 weeks	Allergic diseases development		No statistically significant differences in the incidence of AD	Difference	Di Mauro et al. [115] (2020)
									Treated [18%] versus control [17%]
			60 (30 treated vs. 30 placebo)
									[Fisher’s exact test p = 0.61]
Biological compounds (n = 1)
Open-label, case-control	To determine whether pancreatic enzyme supplementation is an effective and safe treatment in refractory pediatric AD associated with food allergies		Canada	Pancreatic enzyme	1 capsule/day for 6 weeks	AD severity		↓ in the mean SCORAD scores	SCORAD	Singer et al. [116] (2019)
			22 (11 cases vs. 11 controls)						Before [52.3±5.5] versus after [34.6±7.6] with [p = 0.0008]
Diet-related approach (n = 5)
Open trial	To investigate the effect of consuming MSG in processed foods on serum ECP levels among children with AD		South Korea	MSG in processed foods	1 week of processed food avoidance	AD severity		↓ in the mean SCORAD scores and serum ECP levels	SCORAD	Lee et al. [117] (2011)
									Before [23.53±11.68] versus after [12.4±6.01] with [p = 0.046]
									Serum ECP (μg/mL)
			13 (6 processed food-restrict group vs. 7 general diet group)			Serum eosinophil cationic protein		Decrement in serum ECP level was significant correlated with decreased MSG intake	Before [25.33±10.3] versus after [10.25±8.09]
									[p = 0.028]
									Correlations
									Serum ECP and MSG intake
									[r = 0.629; p = 0.028]
Open trial	To describe the treatment and follow-up of 63 children with AD who received a diet at home in which all, but six foods were eliminated for a period of 6 weeks		Manchester, England	Lamb, potato, rice, rice krispies, carrot, and pear	Only consumed 6 foods for 6 weeks	AD severity		52% patients obtained ≥20% improvement in disease severity	Disease severity score	Devlin et al. [118] (1991)
			63						Pre-diet [median: 70; range: 20–240] versus 6 weeks [median: 20; range: 4–180] versus 6 months [median: 10; range: 0–140] with [p = 0.02]
Randomized, crossover	To assess the clinical severity of AD in children after certain dietary modifications and to correlate AEC with dietary modification		North Malabar, India	Egg and cow’s milk	Egg and cow’s milk exclusion diet for 9 weeks	AD severity		Avoidance diet did not show a beneficial effect on AD	SCORAD	Dhar et al. [119] (2019)
			30						Before [18.3] versus after [14.3] with [p = 0.165]
Randomized, single-blind, controlled, parallel trial	To study the short-term effect of a few foods’ diets on AD		Manchester, England	Whey or casein	Diet supplemented with whey or casein hydrolyzate formula for 6 weeks	AD severity and symptoms		No statistically significant improvements in AD symptoms between control and treated groups	Differences	Mabin et al. [120] (1995)
									Body surface area affected
									[Kruskal-Wallis, H = 3.47, df = 2, p = 0.18]
									Skin severity score
				Hydrolyzate formula					[Kruskal-Wallis, H = 2.31, df = 2, p = 0.32]
									Day-time itch score
									[Kruskal-Wallis, H = 1.48, df = 2, p = 0.48]
									Sleep disturbance score
									[Kruskal-Wallis, H = 1.49, df = 2, p = 0.47]
Single-blind, controlled	To ascertain the relationship between the effect of egg exclusion on infantile AD, and egg allergy		Japan	Eggs	Exclusion diet for 2 weeks	Severity of skin symptoms		No statistically significant difference between egg avoidance group and cases with severe skin symptoms	χ2	Aoki et al. [121] (1992)
			138						[p > 0.05]

Targeted group (pediatric, n = 53). AD, atopic dermatitis; AEC, absolute eosinophil count; ARA, arachidonic acid; camp/LL-37, cathelicidin antimicrobial peptide; CCL17, CC chemokine ligand 17; CD, clusters of differentiation; CFU, colony-forming unit; CIs, confidence intervals; DFI, dermatitis family impact; DHA, docosahexaenoic acid; EASI, eczema area and severity index; ECP, eosinophil cationic protein; EPX, eosinophil protein X; FOS, fructo-oligosaccharide; Foxp3⁺, forkhead box P3; GI, gastrointestinal; GLA, γ-linolenic acid; GOS, galacto-oligosaccharide; IDQOL, infants’ dermatitis quality of life; IFN-γ, interferon-gamma; IGA, investigator’s global assessment; IL, interleukin; IUs, international units; MSG, monosodium glutamate; OR, odds ratio; SCORAD, SCORing atopic dermatitis; SIgA, secretory immunoglobulin A; SNP, single nucleotide polymorphism; SPT, skin prick test; TARC, thymus and activation-regulated chemokine; TEWL, transepidermal water loss; TSLP, thymic stromal lymphopoietin; VDR, vitamin D receptor.

Identified Gaps and Future Research Strategies

Finally, this study revealed several critical gaps in the existing literature on AD and dietary intervention (Fig. 4). First, heterogeneity was evident in both the diversity of dietary interventions employed, spanning various food types, study designs, sample sizes, and populations, and the multitude of reported outcomes including skin parameters, cytokines, quality of life measures, and microbiota composition. Second, there was a scarcity of research focusing on adult population, with a predominant focus on pediatrics and pregnant women cohorts. Additionally, there was a lack of alignment between studied dietary interventions and established dietary guidelines. Insufficient consideration of cultural and regional dietary practices was another notable gap with many studies overlooked the influence of diverse cultural and regional diets on the effectiveness and relevance of dietary interventions. Lastly, most studies exhibited a short-term focus, limiting insights into the long-term effects and sustainability of dietary interventions for AD.

Fig. 4.

A summary detailing the identified research gap and proposed future research strategies for AD and dietary intervention.

Discussion

This review analyzed 104 full-text articles assessing dietary interventions for AD in human populations among various age groups, covering publications from 1990 to May 2023. There was a predominant focus on the pediatric population, underscoring an emerging paradigm emphasizing early-life interventions. Most studies originate from Europe and explore diverse dietary strategies. The majority of these studies demonstrated an overall effectiveness in ameliorating AD outcomes without serious adverse effects, affirming the potential of dietary interventions in AD management. Moreover, the increasing number of dietary intervention studies on AD over the years highlights a growing interest and sustained research efforts in this domain.

The emphasis on early-life dietary management signifies a fundamental paradigm shift from conventional symptom-centric approaches toward a proactive strategy aimed at not only managing but potentially preventing AD. Early childhood and infancy are critical developmental stages where targeted nutritional interventions can significantly influence immune system development, gut microbiota composition, and overall long-term health [122]. The evidence in our review suggests the potential of early dietary supplementation, especially prebiotics and probiotics, in pediatrics to positively influence the immune system. For instance, prebiotic supplementation during early infancy with GOS and FOS has been associated with an improved gut microbiota profile by promoting bifidobacterial proliferation [101, 110] and immunomodulatory potential by reducing plasma levels of immunoglobulins, including IgE, IgG1, IgG2, and IgG3 [111]. Probiotic supplementation with Lactobacillus species during early childhood leads to varying outcomes; however, the majority of studies consistently supported their potential immunomodulatory effects. The various effects include the reduced proportion of IgA/IgM-secreting cells [85], lowered expression of various pro-inflammatory cytokines such as IL-4, and IFN-γ [77, 97], and lowered total serum IgE [92, 111]. While early intervention in reducing the risk and severity of AD is promising, it also necessitates a careful and evidence-based approach to avoid unwarranted or ineffective dietary modifications. There is a need for dietary interventions to be planned meticulously to ensure basic nutritional requirements are adequately met, especially in the critical developmental stage of young children. On the other hand, the clinical impact of probiotics on the quality of life for adult patients with AD warrants nuanced consideration. Although there were statistically significant improvements in DLQI scores [37, 38], it is essential to contextualize these findings within the recognized clinically important difference [124, 125]. Therefore, it needs additional large-scale studies that are designed with careful consideration of probiotic composition and study parameters to offer a more nuanced perspective on the clinical relevance of observed improvements in DLQI scores.

The disparity in the number of studies focusing on the adult population compared to the significant emphasis on pediatric populations may be a reflection of historical research biases [126]. However, this emphasis also underscored the need for a more balanced approach and increased attention to adult populations in research related to AD. AD affects both children and adults and understanding its manifestations, triggers, and management in adulthood is equally critical. Adults have distinct lifestyle factors, additional comorbidities, and health concerns that influence AD development and progression [127]. Interestingly, our findings revealed a prevalence of vitamin supplementation for improving AD conditions among adults. Consistent with a recent review on vitamin D [128], we also found that vitamin D supplementation is potentially a viable and well-tolerated therapeutic option to alleviate AD symptoms. This aligns with other studies that highlight the roles of vitamins in immune function, skin integrity, and inflammatory responses [129–131]. Most studies, particularly in the adult population, have commonly utilized a supplementation dose of 1,600 IU of vitamin D3/day. However, we should never extrapolate these findings to pediatric AD patients. The determination of the optimal dose and duration of vitamin D supplementation for children with AD remains a challenging and unresolved issue based on current studies [132]. Furthermore, it is crucial to acknowledge that not all reductions in SCORAD scores by vitamin D supplementation may translate into clinically significant improvements. To bridge the gap between statistical significance and clinical relevance in reported outcomes, future research demands interdisciplinary collaboration involving clinical experts in dermatology, immunology, and nutrition. Establishing a standardized framework, underpinned by methodological rigor, is paramount. This includes well-defined criteria for clinical significance and robust subgroup analyses. These strategies are pivotal for advancing dietary intervention aimed at improving AD across diverse population groups.

There are also several crucial considerations to take into account when designing a robust, widely acceptable, and sustainable RCT for the management of AD. First, dietary preferences and practices vary significantly across different sociocultural contexts [133]. Research often prioritizes standardized dietary interventions rather than acknowledging and studying variations in dietary habits. Thus, understanding the cultural traditions, local culinary practices, and the available food resources is crucial for designing dietary interventions that are well-accepted and sustainable. While various countries and health organizations provide evidence-based dietary guidelines [134], with notable examples such as Dietary Approaches to Stop Hypertension eating plan [135], Dietary Guidelines for Americans (DGA) [136], and Singapore’s “My Healthy Plate” [137], our review highlights a potential gap that revealed a scarcity of research evaluating the effectiveness of dietary interventions aligned with these established guidelines. Additionally, whole-diet interventions are lacking. These observations underscore the need for further investigation into the impact and efficacy of whole-diet interventions adhering to established dietary recommendations to ensure optimal public health outcomes. Multidisciplinary approaches involving dietitians, dermatologists, and other relevant healthcare professionals can enhance the design and implementation of dietary interventions, ensuring that the dietary plans not only target AD management but also the overall health and well-being of the target group. Finally, most intervention studies have a relatively short-term focus, typically ranging from a few weeks to a few months. This short-term orientation is especially limiting for a chronic cutaneous condition like AD. Moreover, treatment response is restricted to the observed short period and the long-term actual benefits of treatment may be potentially underestimated. Adverse effects, risks, or delayed hazards associated with treatments may also be difficult to detect during the intervention trial. For a more comprehensive understanding of dietary intervention’s influence on the natural course of AD, it is crucial to conduct longitudinal studies that extend into adulthood. This extended timeframe allows researchers to track the long-term effects and efficacy of early dietary interventions. Although it can be resource-intensive and require significant time and financial investments, the insights gained from long-term observations would be invaluable in shaping evidence-based strategies for managing AD from early childhood into adulthood.

Limitation and Conclusion

One of the notable challenges in studying interventions for AD is the significant heterogeneity in the types and methodologies of interventions. The diversity extends to the outcome measures used to evaluate the effectiveness of dietary interventions which some studies focus on clinical outcomes like AD severity scores while others assess immunological markers (e.g., cytokine levels, immune cell responses), microbiome composition, patient-reported outcomes (e.g., pruritus, sleep quality), or even genetic changes. Researchers and funding organizations should collaborate internationally, share data and resources, and establish standardized protocols. This approach accelerates research progress and ensures consistency and reliability in AD studies. While we have included studies that adhered to the established guidelines for defining AD, we recognized the extensive heterogeneity within this condition. AD presents a broad spectrum of severity levels and diverse clinical manifestations which may pose a challenge to the generalizability of dietary intervention studies. The effectiveness of a dietary intervention for a specific AD subtype or severity may not necessarily extend across the entire spectrum of variations. Integrating multi-omics data (genomics, proteomics, metabolomics) in future studies can stratify AD into subtypes with unique molecular features, allowing precise assessment of dietary intervention responses [138]. As this review aims to present a comprehensive overview of dietary interventions associated with AD outcomes, the quality of studies was not assessed by the Newcastle-Ottawa Scale. Hence, there may be personal bias and errors, such as misinterpretation of results. Finally, it is important to acknowledge a potential source of bias where studies from non-English speaking countries might have been inadvertently excluded due to our restriction to English-language articles. This could contribute to an underrepresentation of dietary intervention studies from countries like China, potentially underestimating the number of studies in Asia. A separate review focusing on dietary interventions for AD in Asia can be warranted in the future to enable a comprehensive understanding of region-specific dietary approaches, considering the distinct influences of diverse ethnicities, cultures, eating habits, and geographical regions in Asia.

In conclusion, this review identified several effective dietary interventions for reducing AD severity across various targeted populations including adults, pediatric, and lactating or pregnant women. Of these, there were more studies advocated for early-life interventions, particularly through probiotic supplementation in maternal and pediatric diets. Despite the expanding body of intervention research on AD, we emphasize the importance of multidisciplinary approaches and advocate for future RCTs that not only explore improved methodologies but also consider cultural diversity, region-specific dietary approaches, and complement these efforts with longitudinal studies.

Acknowledgments

We would like to extent our gratitude to all authors who have contributed to the studies that we have reviewed and all participants involved in these studies.

Statement of Ethics

All authors have read and consented to the publication of this manuscript.

Conflict of Interest Statement

F.T.C. reports grants from National University of Singapore, Singapore Ministry of Education Academic Research Fund, Singapore Immunology Network, National Medical Research Council (NMRC) (Singapore), Biomedical Research Council (BMRC) (Singapore), National Research Foundation (NRF) (Singapore), Singapore Food Agency (SFA), and the Agency for Science Technology and Research (A*STAR) (Singapore), during the conduct of the study; and consultancy fees from Sime Darby Technology Centre; First Resources Ltd; Genting Plantation, Olam International, and Syngenta Crop Protection, outside the submitted work. The other authors declare no competing interests.

Funding Sources

F.T.C. has received research support from the National University of Singapore, Singapore Ministry of Education Academic Research Fund, Singapore Immunology Network (SIgN), National Medical Research Council (NMRC) (Singapore), Biomedical Research Council (BMRC) (Singapore), National Research Foundation (NRF) (Singapore), Singapore Food Agency (SFA), and the Agency for Science Technology and Research (A*STAR) (Singapore); Grant No. N-154-000-038-001, R-154-000-191-112, R-154-000-404-112, R-154-000-553-112, R-154-000-565-112, R-154-000-630-112, R-154-000-A08-592, R-154-000-A27-597, R-154-000-A91-592, R-154-000-A95-592, R-154-000-B99-114, SIgN-06-006, SIgN-08-020, NMRC/1150/2008, OFIRG20nov-0033, BMRC/01/1/21/18/077, BMRC/04/1/21/19/315, BMRC/APG2013/108, NRF-MP-2020-0004, SFS_RND_SUFP_001_04, W22W3D0006, H17/01/a0/008, and APG2013/108. F.T.C. has received consulting fees from Sime Darby Technology Centre, First Resources Ltd, Genting Plantation, Olam International and Syngenta Crop Protection, outside the submitted work. All funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author Contributions

F.T.C. conceived and supervised the current review study. J.J.L. and M.H.L. contributed to the study design, data analysis, literature review, and interpretation of data. J.J.L. wrote the manuscript draft. All authors have read and approved the final manuscript for submission.

Funding Statement

F.T.C. has received research support from the National University of Singapore, Singapore Ministry of Education Academic Research Fund, Singapore Immunology Network (SIgN), National Medical Research Council (NMRC) (Singapore), Biomedical Research Council (BMRC) (Singapore), National Research Foundation (NRF) (Singapore), Singapore Food Agency (SFA), and the Agency for Science Technology and Research (A*STAR) (Singapore); Grant No. N-154-000-038-001, R-154-000-191-112, R-154-000-404-112, R-154-000-553-112, R-154-000-565-112, R-154-000-630-112, R-154-000-A08-592, R-154-000-A27-597, R-154-000-A91-592, R-154-000-A95-592, R-154-000-B99-114, SIgN-06-006, SIgN-08-020, NMRC/1150/2008, OFIRG20nov-0033, BMRC/01/1/21/18/077, BMRC/04/1/21/19/315, BMRC/APG2013/108, NRF-MP-2020-0004, SFS_RND_SUFP_001_04, W22W3D0006, H17/01/a0/008, and APG2013/108. F.T.C. has received consulting fees from Sime Darby Technology Centre, First Resources Ltd, Genting Plantation, Olam International and Syngenta Crop Protection, outside the submitted work. All funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability Statement

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon reasonable request. Further inquiries can be directed to the corresponding author.