Introduction

As disruption of the healthy microbiome, referred to as dysbiosis, has been recognized as a risk factor for the development and exacerbation of atopic dermatitis (AD), treatments aimed at restoring balance in the microbiome have been explored. Over the past few decades, numerous studies and subsequent reviews and meta-analyses have been conducted, investigating the use of oral probiotics for both treatment and prevention of AD. Though results are mixed, and study methodologies varied, certain populations appear to benefit from their use. Probiotics are considered generally safe, though they should be used with caution in select populations. More recently, topical preparations have been explored in small studies with varying results.

Oral probiotics

As defined by the International Scientific Association for Probiotics and Prebiotics, probiotics refers to “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host” ( ), while a prebiotic is defined as “a substrate that is selectively utilized by host microorganisms conferring a health benefit” ( ). In short, probiotics are active bacterial cultures, and prebiotics are what feed them. Synbiotics are a combination of probiotics and prebiotics. The term postbiotics refers to metabolic byproducts of live bacteria (either secreted or released through lysis) that, like probiotics, provide a health benefit to the host ( ) ( Fig. 24.1 ).

Prebiotics, probiotics, synbiotics, and post-biotics.

Gut commensals in the microbiome aid human health through a variety of mechanisms, including competitive inhibition of pathogenic organisms, protection of the gut epithelial barrier, assisting in nutrient metabolism, and immunomodulation ( ). For example, intake of macronutrients such as complex carbohydrates are metabolized by gut bacteria through fermentation into end products used by humans for energy, such as short-chain fatty acids (SCFAs) ( ). To be effective, bacterial strains used in oral probiotics should be able to survive passage through the acidic environment and digestive enzymes of the upper gastrointestinal (GI) tract, adhere to intestinal epithelial cells, colonize the lower GI tract, and be safe for use (i.e., nonvirulent, nontoxin producing) ( ).

Dosage of probiotics is measured in colony-forming units (CFU), which indicate the number of live or viable organisms per serving (i.e., capsule or gram). Probiotic dosage typically ranges from 1 to 10 billion CFU, and in guidelines from the World Gastroenterology Organisation (WGO), due to variation in dosage and strains used in studies, there is no one recommended dose; instead, dosage should be based on the studies of the disorder treated ( ). Though probiotics and prebiotics are readily available as supplements in drugstores and supermarkets, several common foods contain live microbes and prebiotics as well ( Table 24.1 ).

Dysbiosis has been purported to play a role in the development and exacerbation of AD, and as such, probiotic use has been explored in both the prevention and treatment of the disease. The exact mechanism of action of probiotics is unclear, though likely through influencing both the microbiome and the immune system ( Fig. 24.2 ). Like commensals, probiotics aid in barrier function, compete for nutrients or displace pathogenic organisms, produce AMPs, and stimulate the immune system to target pathogenic organisms ( ). Lactobacillus , a genus of commensal bacteria in the phylum Firmicutes (a predominant gut phylum), is commonly used in probiotics. Studies have shown a range of benefits by different species within Lactobacillus . For example, Lactobacillus rhamnosus GG can protect the gut barrier by minimizing mucosal lining permeability and protecting against oxidative stress-induced damage to tight junctions, the intercellular connections that prevent the passage of molecules ( ). Another Lactobacillus species, Lactobacillus reuteri , produces reuterin, an antimicrobial compound with broad coverage against gram-positive bacteria, gram-negative bacteria, fungi, yeasts, and protozoa ( ).

Synbiotic and topical probiotic mechanisms. Prebiotics feed beneficial organisms (probiotics) and have an anti-inflammatory effect, while the probiotic organisms compete with pathogenic organisms in the gut, improve barrier function and modulate the immune system. In the skin, increased regulatory T cells decrease the Th2 response elicited from allergen exposure. Topical probiotics applied to skin can limit levels of S. aureus and improve barrier function. AMPs , Antimicrobial peptides. DC , Dendritic cell. IgG , Immunoglobulin G. IgA, Immunoglobulin A. SCFA , Short-chain fatty acids. TReg , Regulatory T cell. Th2 , T helper type 2 cell.

Both the innate and adaptive immune systems are influenced by probiotic administration. Depending on the strain, this is accomplished through different mechanisms. L. reuteri and Lactobacillus casei , for example, induce increased inhibitory regulatory T-cell (Treg) differentiation after detection by tolerogenic dendritic cells, while Lactobacillus acidophilus , Lactobacillus bulgaricus , and Bifidobacterium bifidum increase protective immunoglobulin G (IgG) and IgA antibodies ( ). Regardless of species and strain differences, probiotics can inhibit the differentiation of naïve T cells to T helper type 2 (Th2) cells and stimulate the production of antiinflammatory cytokines such as interferon-gamma and interleukin-10 (IL10) ( ). Prebiotics also play a role by increasing production of SCFAs, byproducts of fermentation by gut commensals, which have antiinflammatory effects and decrease pH, creating a more favorable environment for commensal over pathogenic bacteria and promoting gut homeostasis ( ).

Treatment of AD with oral probiotics

Numerous studies, systematic reviews, and meta-analyses on the use of probiotics for the treatment of AD have been published, with conflicting results and conclusions. The most recent Cochrane review published by analyzed 39 randomized clinical trials (RCT) on the use of probiotics (specifically Lactobacillus and Bifidobacteria ) in 2599 participants with at least mild AD of all ages. Key findings from the review were that the use of probiotics as treatment did not help with AD symptoms or quality of life, concluding that the “use of probiotics for the treatment of eczema is currently not evidence-based” ( ). This review was an update to the 2008 Cochrane review of 12 RCTs with 781 participants (pediatric patients only), which also found lack of benefit with probiotic use as treatment for AD ( ).

A meta-analysis specifically investigating improvements in the scoring of AD (SCORAD) measure following probiotic therapy was conducted by and identified 25 RCTs with 1599 participants. Results from this analysis conflict with the Cochrane reviews and showed a benefit in the use of probiotics in children at least 1 year of age (–5.74-point difference in SCORAD compared to controls) and adults (–9.69-point difference in SCORAD compared to controls), though this benefit was not seen in infants less than 1 year old. Differences in benefit were influenced by type of probiotic and severity of AD, with greater benefits seen in those supplemented with Lactobacillus or a mixture of bacteria rather than Bifidobacterium alone ( ). In subgroup analysis, this benefit from probiotics was limited to those with moderate to severe AD rather than those with mild disease ( ). However, some studies included in the meta-analysis allowed patients to continue using topical emollients and medications in addition to probiotics, making these results challenging to interpret. Similar results were seen in a more recent meta-analysis focused on infantile AD, which identified a significant benefit of probiotic supplementation (specifically with Lactobacillus ) in children ages 1 to 18 years with moderate to severe AD ( ).

The benefits of probiotics also vary dependent upon the population studied. A meta-analysis conducted by found that only certain strains of Lactobacillus improved SCORAD in children, showing that Lactobacillus fermentum and Lactobacillus salivarius (as well as probiotics of varying combinations of Lactobacillus and Bifidobacterium strains) led to significant improvement, while single strain treatment with L. rhamnosus GG or L. plantarum did not. Differences in benefit were also age dependent, and even geography dependent, with significant improvements in SCORAD only seen in children age greater than 1 year and in those in Asia, though not in Europe ( ). This suggests that there may be different phenotypes of AD that respond more favorably to probiotics than others.

Prevention of AD with oral probiotics

Given the potential role the gut microbiome plays in the development of AD, multiple studies have investigated the use of probiotics in the prevention of disease. Conclusions of meta-analyses have largely agreed that, within certain parameters, there is evidence to support the use of probiotics in a preventative role. A large systematic review and meta-analysis by analyzed 28 studies that included 6907 infants and children exposed to probiotics prenatally and/or after birth prior to diagnosis of AD. In this study, significant reduction in the incidence of AD was found when probiotics were given both to the mother in utero and to the infant postnatally, but not if only one or the other. This benefit was only seen when treatment occurred through the age of 6 months, with no decreased risk in the development of AD in those treated for more than 1 year. Contrary to studies on treatment of AD, this benefit of disease prevention was seen across strains, including Bifidobacterium, Propionibacterium , and Lactobacillus strains, with certain Lactobacillus strains ( L. rhamnosus and Lactobacillus paracasei ) showing greater efficacy ( ). This benefit continued to be sustained on long-term, at least 5-year, follow-up ( ). Multiple other systematic reviews and meta-analyses have also concluded that early probiotic use decreases risk of AD development ( ).

The use of prebiotics, given as oligosaccharides (i.e., fructo- and galactooligosaccharide) for prevention of AD has also been evaluated. A Cochrane review published in 2013 by Osborn and Sinn investigated the role of prebiotics in infants for the prevention of allergic disease. A significant risk reduction was observed in the development of AD (four studies, 1218 infants), though not in the development of allergy or asthma ( ). A strength of this review is that there was no significant heterogeneity among AD studies. Contradictory to these findings, a meta-analysis published that same year included subgroup analysis of three RCTs investigating prebiotic use and found no benefit in the prevention of AD, though heterogeneity among the included studies was moderate (I ² of 48%) ( ).

The use of synbiotics for both the prevention and treatment of AD has also been investigated. In a meta-analysis published in 2016, Chang et al. included six studies on the treatment and two studies on the prevention of AD in children and found that use of synbiotics were shown to help in the treatment but not the prevention of AD. In this meta-analysis, significant improvements in SCORAD were seen in those ages 1 to 18 years treated for 8 weeks, and when treated with mixed bacterial strains rather than a single strain. Synbiotics were associated with lower incidence of AD in the two individual studies used in the meta-analysis; however, when combined, the pooled relative risk was not significant ( ). Theoretically, synbiotics should perform as well as or better than either separate component; however, to date, there are no studies specifically comparing probiotics to synbiotics in the prevention and treatment of AD, and comparisons of separate meta-analyses is challenging in the setting of study heterogeneity ( ).

Recommendations

Translating current meta-analyses into evidence-based recommendations is challenging given the heterogeneity among studies, particularly in strains used, doses used, timing and duration of supplementation, concomitant treatments, and outcome assessment. Though recognizing the low quality of evidence, the World Allergy Organization (WAO) currently recommends considering supplementation in the following scenarios:

a.“using probiotics in pregnant women at high risk for having an allergic child;

b.using probiotics in women who breastfeed infants at high risk of developing allergy; and

c.using probiotics in infants at high risk of developing allergy” ( ).

The WAO also recommends use of prebiotics specifically in “not-exclusively breastfed infants” ( ), though no specific strains or regimens are provided. A follow-up meta-analysis to these recommendations, specifically evaluating use of L. rhamnosus GG (LGG) in the abovementioned three groups, found that LGG supplementation did not reduce the risk of AD ( ). In their guidelines on the management of AD, the American Academy of Dermatology (AAD) acknowledges the current interest in the study of probiotics but does not currently recommend the use of probiotics or prebiotics for treatment of established AD based on the available evidence ( ). Other expert groups have not endorsed the use of probiotics, prebiotics, or synbiotics for the prevention of allergic disease, including the American Academy of Pediatrics, National Institute of Allergy and Infectious Diseases, European Academy of Allergy and Clinical Immunology, European Society of Paediatric Gastroenterology, Hepatology, and Nutrition, and the Food and Agriculture Organization of the United Nations/World Health Organization ( ).

Though oral probiotics have been studied in other cutaneous diseases as well, such as psoriasis ( ) and acne ( ), many questions remain regarding their mechanism of action. In addition to unanswered questions already discussed, there are questions regarding how long probiotics need to stay in the gut to work, how many CFU are needed to make a compositional change, and whether a transient encounter in the gut is enough to change the immune milieu or behavior of the microbial community.

Topical microbial therapy

As skin dysbiosis, particularly overgrowth of Staphylococcus aureus , is associated with AD, topical approaches to return the skin to a balanced state have been studied. Although topical antibiotics in conjunction with topical corticosteroids can decrease S. aureus load, there is no clear evidence supporting the use of topical antibiotics in AD for improvement in symptoms in the absence of infection ( ). Routine topical antibiotic use is not currently endorsed by the AAD ( ). Topical microbial approaches that have recently been studied include topical probiotics (live organisms), bacterial lysates (products of lysed bacterial cells), endolysins, and clothing ( Table 24.2 ). As with oral probiotics, topical strains should possess select characteristics to return the skin microbiome to balance, including the ability to adhere to skin, decrease adhesion of potential skin pathogens, compete against pathogens, decrease biofilm formation by pathogens, and in some cases even break down mature biofilms ( ). Topical autologous microbial transplant using the patient’s own protective commensals has also been studied ( ).

Lactobacillus

Lactobacillus johnsonii (even once heat-treated and no longer able to replicate) has been shown to stimulate host innate immune response, with expression of antimicrobial peptides (AMPs) such as cathelicidin and β-defensin ( ). studied the effects of a lotion containing a strain of heat-treated L. johnsonii on 21 patients with mild to moderate AD and positive swabs of S. aureus at the start of the study. Participants applied the lotion to a target lesion twice daily for 3 weeks and reevaluated using SCORAD. Results showed a significant decrease of S. aureus in treated skin, with significant clinical improvement in SCORAD, in a well-tolerated lotion ( ).

Topical use of L. reuteri has also been investigated, though not yet in human in vivo application. conducted in vitro and ex vivo studies on human epidermal explants evaluating both live L. reuteri and lysate products of the bacteria. Both the live bacteria and bacterial lysate reduced inflammatory cytokines IL6 and IL8, and both improved skin barrier function through multiple mechanisms, including upregulation of aquaporin-3 (live bacteria) and laminin A/B expression (lysate). Microbial inhibition was also observed with use of the live culture, but not the lysate, with increased antimicrobial activity against pathogens Pseudomonas aeruginosa and Streptococcus pyogenes compared to commensals Staphylococcus epidermidis and Cutibacterium acnes ( ).

Roseomonas

investigated culturable gram-negative (CGN) bacteria from patients with AD (N = 17) and healthy volunteers (N = 26), and found that Roseomonas mucosa was the predominant species in the healthy volunteers. CGNs from healthy controls can limit S. aureus growth, improve barrier function through reduced transepidermal water loss, and induce an innate immune response ( ). In a single arm follow-up study, 15 patients with AD were treated with a topical solution containing R. mucosa (10 adults treated twice weekly for 6 weeks; 5 pediatric patients treated twice weekly for 16 weeks) ( ). Combined, 10 of the 15 patients achieved a greater than 50% improvement in SCORAD, with no reported adverse events. Lack of response was associated with a family history of AD that continued through adulthood.

Vitreoscilla

Vitreoscilla filiformis is a filamentous gram-negative bacterium found in thermal springs. It has been shown to have immune-modulating effects through induction of dendritic cells, production of IL10 (an antiinflammatory cytokine), and subsequent priming of regulatory T cells (Tr1 cells) ( ). Two double-blind, randomized, placebo-controlled studies investigating its use topically, one with participants with mild AD ( ), the other with participants with moderate AD ( ), found significant decreases in SCORAD in the treated group after 4 weeks, compared to the vehicle-only control group. Changes in participant microbiome were also noted in both studies. observed a nonsignificant decrease in bacterial colonization in the treated group. observed a significant increase in Xanthomonas in the treated group, while the control group had a nonsignificant increase in Staphylococcus . Additional human studies have also observed clinical improvement in AD with topical application of Vitreoscilla ( ).

Bifidobacterium

A cream containing the lysate of Bifidobacterium longum , a gram-positive anaerobe, was tested by in a randomized, double-blind, placebo-controlled trial of 66 female participants with reactive skin, as determined by report of sensitive skin, leg dryness, and face roughness. Patients were instructed to apply either the lysate cream (N = 33) or control cream (N = 33) twice daily for 2 months, with evaluation at baseline, 1-month, and 2-month follow-up, assessing skin sensitivity, skin barrier function, and skin hydration. A decrease in skin sensitivity at 2 months, improvement in skin barrier function at 2 months, and a reduction in facial roughness and leg dryness at 1 month in the study group were observed. Although not tested in AD, authors suggest that topical B. longum can improve skin barrier function ( ).

Staphylococcus

Skin dysbiosis in patients with AD is marked by changes in the abundance of different species of Staphylococcus bacteria. Select species of coagulase-negative Staphylococcus (CoNS) are known to produce AMPs that limit the growth of S. aureus . used this concept to test autologous transplant of CoNS onto five S. aureus– positive participants with AD. In this double-blind study, a cream containing antimicrobial strains of S. epidermidis and/or Staphylococcus hominis isolated from each of the five patients was applied back to the originating subject on one forearm, while vehicle was applied to the other forearm as control. Assessment of bacterial composition 1 day after application showed a significant decrease in S. aureus abundance on the area that received treatment with the antimicrobial CoNS, with no change in the abundance of S. aureus on untreated and vehicle areas ( ). Clinical assessment of AD was not measured in this study, but results indicate a potential effective topical approach to treating skin dysbiosis. Additional investigation into autologous transplant is still underway ( ).

Streptococcus

Streptococcus thermophilus is a gram-positive facultative anaerobe that has been shown to increase ceramide levels both in vitro and in vivo ( ). On the premise that topical application of ceramides can improve barrier function, and thus potentially benefit patients with AD, investigated use of a cream containing S. thermophilus on 11 patients with AD, with participants treating one forearm with the formulation and the other with the vehicle as control. Evaluation of participants in this study at 2 weeks showed significant improvement in S. thermophilus -treated areas with regard to erythema, scaling, pruritus, and vesiculation. A statistically significant increase was also seen in the ceramide levels of the stratum corneum of treated skin, further supporting the importance of having a functioning skin barrier in control of AD. It should be noted that all study participants stopped application with the vehicle within 5 days due to lack of efficacy ( ).

Endolysins

Endolysins are enzymes produced by bacteriophages (viruses that infect bacteria) that break down the peptidoglycan cell wall of the bacteria, resulting in cell lysis ( ). Staphefekt SA.100 is a recombinant phage endolysin that specifically targets S. aureus , including both methicillin-sensitive and resistant (MRSA) strains ( ). Its use as a targeted topical therapy was theorized to decrease S. aureus levels without indiscriminately affecting skin commensals, as would be seen with a topical antibiotic or antiseptic ( ). However, such efficacy was not demonstrated in clinical application. In a recent double-blind, vehicle-controlled study, 100 adult patients with clinically noninfected moderate to severe AD were given either Staphefekt SA.100 or vehicle to apply topically twice daily for 12 weeks ( ). The primary outcome was the difference in topical corticosteroid use with patients using the topical endolysin versus placebo. At the end of the 12-week intervention, no statistically significant difference in use of topical corticosteroids was observed between the two groups, nor were significant reductions in S . aureus seen, as measured by culture and qPCR ( ).

Antimicrobial-embedded clothing

The use of chitosan, a biopolymer found in nature as a component of the exoskeleton of shellfish, has been explored in multiple products, including cosmetics, medical products, and agriculture, due to its antimicrobial activity ( ). Its use in clothing has also been studied and shown to have antimicrobial activity against S. aureus ( ). In a study of 78 adults with AD, the use of chitosan-coated pajamas improved SCORAD values from patient baseline; however, this improvement was not statistically significant when compared to controls ( ). A systematic review of 13 studies investigating the use of other antimicrobial clothing items (silver-coated, silk, borage oil) found that these were generally safe and may improve disease severity ( ).

Safety and Food and Drug Administration regulations

Use of probiotics is considered generally safe and well tolerated. A systematic review and meta-analysis evaluating the safety of Lactobacillus and Bifidobacterium in pregnancy (eight RCTs, 1546 patients) found that their use had no effect on cesarean section rate, birthweight, or gestational age ( ). The most recent Cochrane review (see Oral Probiotics, earlier) included this review as well as two other reviews of probiotic safety and found no increased risk or adverse effects with their use ( ). The meta-analysis conducted by on the use of probiotics in treating AD reviewed safety data in 9 of the 25 included randomized trials and found no significant differences in adverse events in treatment versus control groups, with GI symptoms being the most common. Thus probiotic use in the treatment and prevention of AD is generally considered safe.

However, caution may be prudent in specific patient populations. There have been reports of serious adverse events such as sepsis, fungemia, and bowel ischemia in at-risk patients, including the critically ill in the intensive care unit, critically sick infants, postoperative and hospitalized patients, and immunocompromised patients ( ). There is also emerging concern that probiotics may alter the metabolism of anticancer drugs, altering their effects ( ).

The US Food and Drug Administration (FDA) allows probiotics to be regulated as a dietary supplement, food, or drug, which is dependent upon the intended use ( ). Food is defined as “articles used for food or drink for man or other animals,” while a dietary supplement is defined as “a product intended to supplement the diet” and includes vitamins, minerals, herbs, and other nonconventional food items intended for ingestion ( ). To be categorized as a drug, the intended use of the product is for “diagnosis, cure, mitigation, treatment, or prevention of disease” ( ). Therefore, for a probiotic to market health benefits like a drug, it would need to go through the FDA approval process. If a probiotic is categorized as a food, it needs to be generally recognized as safe (GRAS), and if not, then it would need to have premarket clearance. A dietary supplement can go direct to market, and even make claims regarding its effect on a body structure or function (i.e., supports the immune system), provided it does not make specific health claims as related to a disease, which would then change its intended use to that of a drug ( ). Currently, a variety of probiotics have been categorized as dietary supplements and foods; however, to date, none have been approved to prevent or treat health problems under the category of drug ( ).

Treatments on the horizon

Studies on the use of oral probiotics for the treatment and prevention of AD have shown mixed results. Though conclusions in meta-analyses are limited by study heterogeneity, investigators are now also examining how differences in individual microbiomes play a role in their effectiveness. studied gut microbiome changes following administration of an 11-strain probiotic (predominantly Lactobacillus and Bifidobacterium strains) and found that the gut microbiome of study participants varied at baseline, and subsequently only a portion were colonized by the administered strains (“permissive”), while others were not (“resistant”), as measured by samples collected via colonoscopy and deep enteroscopy (representing gut mucosal colonization) following 3 weeks of treatment. This was independent from those bacteria shed in participant stool ( ). In a separate study by , changes in the gut microbiome were investigated following oral antibiotics, with participant intervention being probiotics, autologous fecal microbiome transplant (aFMT), or no intervention (“watchful waiting”). Results showed that the gut microbiome of those receiving aFMT returned to baseline in as quickly as 1 day, while those taking probiotics did not return to baseline by 5 months after completion of probiotics course; those in the watchful waiting group returned to baseline at a time in between interventions (around 3 weeks) ( ).

Microbiome precision editing, or modifying a person’s microbiome for treatment of disease, is also being explored as a future investigative direction ( ). studied the impact of tungsten administration (a metal that can inactivate an Enterobacteriacea cofactor) on mice with chemically induced colitis. In the study, mice were colonized with bacteria ( Escherichia coli , Family: Enterobacteriaceae ) from human patients with inflammatory bowel disease, treated with oral tungstate, and reported to have a selective decrease of Enterobacteriaceae with subsequent decrease in inflammation, without observed changes to the microbiome in homeostasis ( ). Together, these examples suggest that the effects of probiotics vary by person, dependent on individual microbiomes, and that a tailored approach may produce better results.

New insights into the impact on the microbiome of already existing treatments are emerging. A recent investigation into the effects of dupilumab, an IL4a receptor antagonist that is FDA approved for the treatment of moderate to severe AD, demonstrated that treatment with dupilumab not only resulted in clinical improvement but also in microbiome changes ( ). Prior to treatment, higher levels of S. aureus and lower microbial diversity were seen in lesional skin, consistent with other studies. Following 16 weeks of treatment with dupilumab, significant decreases in S. aureus and increases in microbial diversity were observed ( ). Further investigation into topical therapies that alter the microbiome and lead to improvement in AD are also underway. The safety and efficacy of a targeted microbiome transplant (TMT) lotion on adult patients with moderate to severe AD and positive S. aureus skin lesions is ongoing ( ).

Summary

Interest in the use of probiotics for the prevention and treatment of AD has increased as our understanding of the microbiome has expanded. However, evidence gathered on the benefits of use of oral probiotics from systematic reviews and meta-analyses has been less convincing. As of now, a promising new avenue for treatment is the topical use of products geared toward altering the cutaneous microbiome. However, studies to date remain small and limited.

In the United States, currently no probiotics have gone through the rigorous testing for approval as a drug, and so cannot be sold for the purpose of treatment or prevention of disease. Thus they remain largely unregulated in regard to strain, dosage, and validity of health claims. Recognizing that the microbiome of each individual is unique and may contribute to the heterogenous clinical response to treatment with probiotics, future work may investigate and utilize a more personalized approach to treatment.