Probiotics improve atopic dermatitis in children and adolescents: a double blind, placebo-controlled study.

,


Introduction
Atopic dermatitis (AD) is a chronic predominantly type 2 inflammatory skin disease characterized by intense pruritus and associated with atopic comorbidities 1 .
AD diagnosis is based on clinical symptoms and Scoring for Atopic Dermatitis index (SCORAD) helps physicians to assess severity on a regular basis 2 .
Several studies have demonstrated a close relationship between microbial dysbiosis and development of allergy during childhood 3 .The composition of the gut microbiome, including the presence of Bifidobacterium andLactobacillus species, is important for health-promoting effects 4 and probiotics, live microorganisms whose adequate administration benefits the host's health 5 , have been attractive in the treatment of allergic diseases 6 .
This study aimed to evaluate the clinical benefit of combined probiotics in the management of children and adolescents with AD.In addition, other assessed parameters were sensitization (serum IgE levels and skin prick test), inflammation (IFN-g, IL-1β, IL-4, IL-6, IL-8, IL-17, TNF-α), and tolerance (IL-10 and TGF-β).

Study design (Figure 1)
This was a double-blind, randomized and placebo-controlled clinical trial involving children and adolescents aged between 6 months to 19 years with mild, moderate or severe AD, diagnosed according to the Hani-fin&Hajka criteria 7 .Consecutive patients were invited to participate and then they were clinically assessed for AD severity (on the basis of SCORAD) and submitted to laboratorial analysis and to skin prick test (SPT).Patients were stratified in mild, moderate, or high severity groups.Blocks of four (04) patients were considered in each severity group for random allocation, according to a list of computer-generated random numbers.Each group had the same amount of patients receiving placebo or probiotics.Patients were enrolled until August 2015, when they started the treatment, and they were followed up for one year.The recruitment, randomization, clinical assessments, laboratory analysis and SPT were performed by the same investigator.

Patient population
Patients followed up at the Pediatric Allergy Outpatient Clinics in a tertiary hospital and that presented at least one documented flare were included in the study.Patients were excluded if they had other skin diseases, which could interfere; oral corticosteroid or immunosuppressant use within 30 days; lose over 50% of the assessments conducted along the trial.

Sample size
The determination of the sample size required for the study was based on the SCORAD mean comparison at the end of the follow-up, between probiotics and placebo groups, considering a parallel group trial.Establishing a variance of 10.2 points and a clinically relevance difference of 8.7 points, as well as the significance level of 0.05 and statistical power of 0.90, required a total of 60 patients to conduct the study.

Intervention
A formula containing a mixture of 4 probiotics strains (Probiatop®):Lactobacillus rhamnosus HN001 -10 9 Colony Forming Units (CFU); Lactobacillus acidophilus NCFM -10 9 CFU; Lactobacillus paracasei Lcp-37 -10 9 CFU; and Bifidobacterium lactis HN019 -10 9 CFU, was used in the dose of one gram (one sachet) per day.Placebo consisted of one gram of maltodextrin, in the powder form, which was administered as a sachet to ensure similarity to the probiotic.Probiotic and placebo sachets were identical, except for their identification as group A or B, to guarantee that participants and investigator were blind to their contents.Patients were instructed to dilute one sachet in 100 mL of natural water and to take it every morning on a daily basis.Treatment lasted the first six months of the trial.A committee composed of three physicians who were not involved in the trial ensured external monitoring.

Assessments
Clinical assessments were conducted every three months and blood samples for laboratory analysis were collected before the trial start, after the treatment and at the end of the follow up.The skin prick tests were carried out at the end of the follow up.
For cytokines analysis, peripheral blood was collected and serum was obtained by centrifugation at 2500 rpm and 4ºC for 10 min.The serum samples were stored and frozen at -80degC until all analyses were performed.In the end, the wells of a 96-well plate were covered and incubated with purified antibodies anti-IL-17 (R&D System, Minneapolis, MN, EUA), anti-IFN-γ, anti-IL-1β, anti-IL-4, anti-IL-6, anti-IL-8, anti-IL-10, anti-TNF-α, or anti-TGF-β (BD Bioscience, San Diego, CA, EUA) and enzyme-linked immune sorbent assay (ELISA) was performed according to the manufacturer's instructions.
Serum total IgE levels were measured by a fluoroenzyme immunoassay (Phadia ImmunoCap System, Uppsala, Sweden), according to the manufacturer's instructions.The results were expressed in kU/L and were considered elevated when they were higher than 100 kU/L.Skin prick test was performed in all patients using a panel of standardized extracts (Greer®): mites (Dermatophagoides pteronyssinus , Dermatophagoides farinae and Blomia tropicalis ); cockroachs (Blatella germanica andPeriplaneta americana ); pet's dander (cat and dog) and some food allergens (milk, egg, soybean, wheat, peanuts, seafood and fish).The test was considered positive when the diameter of the wheal was greater than 3 mm emerged 20 min after application.

Ethics considerations
The study (ClinicalTrials.gov#NCT02519556) was approved by the Ethics Committee of Hospital das Clínicas de Ribeirão Preto (University of São Paulo).Parents provided a written informed consent for their child s participation; adolescents signed the assent term.

Statistical analysis
The SCORAD was represented as percentage.The values observed at times 3, 6, 9, and 12 months were divided by the value observed at T0 and then multiplied by 100.Data analysis was made by adjusting simple and multiple linear mixed models, that considers the dependence between measurements for the same individual along the trial.
Statistical analysis of the variable serum IgE was similar than performed for SCORAD.
To verify the association of the intervention groups with the cytokines and the SPT, log-binomial regression models were proposed and raw and adjusted relative risks and their 95% confidence intervals for the categories improvement and worsening of symptoms.
All statistical analysis was performed with software SAS 9.4.

Baseline description of patients
Sixty patients were recruited and randomized in two groups (Placebo and Probiotics) and twenty patients were discontinued (Figure 2).Table 1 shows the characteristics of both groups after recruitment phase.
Characteristics of patients at baseline were demonstrated in the Table 2. Mild SCORAD was between 5.5 and 23.9 (mean 14.7), moderate SCORAD was between 25.4 and 49.8 (mean 37.6) and severe SCORAD was between 50.3 and 85.5 (mean 67.9).
The improvement in SCORAD persisted three months more after treatment has been discontinued (T9), even after adjustment for the co-variables (p = 0.04; CI 95%, 0.78-27.70),when compared to the probiotic group itself.However, the clinical assessment between T9 and T12 showed a significant increase in SCORAD (p = 0.04; CI 95%, -28.71-0.68) in probiotic group.

Sensitization assessment
Regarding serum IgE levels, the groups did not differ significantly (Table 3).Regarding to SPT, relative risk was not significantly different between the groups.

Tolerance and inflammation assessment
Regarding to tolerogenic and inflammatory cytokines, relative risk was not significantly different between the groups.

Discussion
This randomized, double-blind and placebo-controlled trial evaluated the clinical and laboratorial efficacy of a mixture of probiotics containingLactobacillus rhamnosus , Lactobacillus acidophilus ,Lactobacillus paracasei , and Bifidobacterium lactis in children and adolescents with AD.After treatment for six months, SCORAD decreased in the probiotic group as compared to the placebo group (p = 0.03), even after adjustment for co-variables (p = 0.02).However, the groups did not differ regarding to inflammatory or tolerogenic cytokines, SPT, or serum IgE levels.To our knowledge, this is the first Brazilian randomized, double-blind and placebo-controlled trial that has shown clinical benefits of probiotics in patients with AD.
Allergic diseases prevalence has increased in recent times, in both high and low-resource countries 8,9 .Because a high morbidity of AD 10 , it is necessary to find an accessible, safe, and effective treatment, which can minimize and/or override the skin inflammation long-term effects.There is growing evidence that the relation between microbiota and immune response can influence in this disease 11 .In general, combined administration of probiotics (before and after birth) has been beneficial, and Lactobacillus rhamnosus GG andBifidobacterium seen to be the most efficient strains 12 .
Han et al. 13 performed a randomized, double-blind, placebo-controlled study lasted twelve weeks to evaluate the influence of Lactobacillus plantarum CJLP133 in children with AD aged between 1 and 12 years.The authors showed SCORAD improvement and a significant decrease in IFN-γ, IL-4 and eosinophil's number.In other study, Wang et al. 14 evaluated the effect ofLactobacillus paracasei alone, Lactobacillus fermentumalone, and the association of both strains in AD patients.Four months after the treatment was discontinued, the authors observed reduction of SCORAD in all groups that received probiotics as compared to the placebo group.Huang et al. 15 carried out a recent meta-analysis including thirteen randomized clinical assays (n = 1070 patients) and demonstrated that probiotics may have the potential to decrease SCORAD values in children with AD.Recently, Navarro-Lopez et.al. 16 published a double-blind, placebo-controlled clinical trial conducted with children and adolescents between 4 and 17 years of age with moderate AD.Patients were supplemented daily with a mixture of probiotics composed by Bifidobacterium lactis CECT 8145, Bifidobacterium longum CECT 7347, and Lactobacillus casei CECT 9104 for 12 weeks.The authors showed reduction in SCORAD and use of topical corticosteroids.
Our study is in accordance with meta-analysis of Huang et al. 15 and with the other aforementioned doubleblind, placebo-controlled studies.We found a significant reduction of SCORAD after treatment with a mixture of probiotics (Lactobacillus andBifidobacterium ) for six months as compared to the placebo group.The reduction of SCORAD persisted until three months after treatment was discontinued, suggesting a short-term beneficial effect.Probably this clinical beneficial effect did not persist until the end of the study because treatment was stopped shortly.Other possibility is patient's microbiota has been already developed by various host-and environment-related factors.Some studies demonstrate that around three years old, the microbiota composition resembles the microbiota composition in adults 17 .
Two main biological pathways are involved in pathogenesis of AD: epidermal epithelial dysfunction and modifications in innate/adaptive immune response 11 .T helper cells are particularly important in inflammation; they determine the immune response degree and direction.T regulatory cells (Treg) are essential to limit excessive immune-inflammatory response.Reduced Treg number in early life is a risk factor for later AD development 10 .In the acute phase of AD, Th2 cell-mediated immune response triggers the inflammatory process.On the other hand, Th1 cell-mediated immune response predominates in the chronic phase of the disease.Some evidences have suggested that Th17 cells are also involved in AD pathogenesis 18 .Probiotic bacteria can affect dendritic cells and modify the T helper cells balance 10 .Previous studies have shown that probiotics inhibit dendritic cell maturation and hence decrease naive T cells differentiation into Th2 cells 19,20 .Kim et al. 21,22 demonstrated that probiotics reduce allergic disease expression in mice by acting in dendritic cells.Holowacz et al. 23 showed that a mixture of probiotics reduced significantly chronic inflammation in rat skin, as judged from the lower levels of pro-inflammatory cytokines (IL-1β, IL-6, TNF-α, IL-17, and IL-22) and higher levels of tolerogenic cytokines, such as IL-10.However, results concerning the effect of probiotics on T cells subgroups and cytokines in children with AD are conflicting.Some studies have reported a tendency toward the Th1 profile resulting in increased inter-feron gamma (IFNγ) production, whereas other studies have not detected any effect on Th1/Th2 balance or Treg 10 .Yeşilova et al. 24 performed a double-blind, randomized, placebo-controlled clinical assay to investigate how eight-week treatment with a combination of probiotics (Bifidobacterium bifidum, Lactobacillus acidophilus, Lactobacillus casei and Lactobacillus salivarius ) affected children with AD aged between 1 and 13 years old.They concluded that the probiotics effectively reduced SCORAD and serum levels of IL-5, IL-6, IFN-γ and IgE, while IL-2, IL-4, TNF-α, and IL-10 remained unaltered.On the other hand, Ludwing et al. 25 verified thatLactobacillus rhamnosus GG soluble mediators (LSM) did not modify the number or the maturation stage of dendritic cells in healthy donors; nevertheless, these cultivated cells induced IFN-γ and IL-2 production in TCD4+ and CD25+ cells.
In our study, probiotics and placebo groups did not differ in acting on inflammatory (IFN-γ, IL-1β, IL-4, IL-6, IL-8, IL-17, and TNF-α) or tolerogenic (IL-10 and TGF-β) cytokines.Although we did not carry out cell culture, our evaluation included a panel of cytokines that represent all possible immune responses (Th1, Th2, Th17, and Treg), which led us to infer that probiotics did not interfere the cell subtypes.We can point possible reasons for these results.Yeşilova et al. 24 used a higher dose of probiotics than we did (2 g x 1 g).Besides that, they analyzed cytokines in the plasma whereas we analyzed cytokines in the serum.Nevertheless, studies analyzing cytokines in humans with AD and treated with probiotics are scarce, which prevents us from drawing more conclusions about other possible factors influencing our results.
Similarly to cytokines, probiotics and placebo groups did not differ in acting on serum IgE levels or SPT results.We did not find studies that evaluated SPT in AD patients treated with probiotics.
Probiotics, especially lactobacilli and bifidobacteria 26 , are known to be safe 27 .The adverse effects most commonly described are diarrhea, vomiting, and increased flatulence 27 .However, invasive infections have been observed in immunocompromised adults, but are rare in breastfed infants 26 .We did not verify any adverse effect in the group of patients that used probiotics.
Strengths of this study are that no clinical trials explored the effect of Bifidobacteria in combination with Lactobacillus in children.Moreover, the safety profile of probiotic treatment was not adequately studied, accordingly to a recent systematic review and meta-analysis performed by Zhao et al. 28 The main limitations of the present study include the small sample size and the difficult of patients' adhesion.These difficulties could be justified by the participants' age range.Parents usually feel insecure and question the need to subject their children to the study.Furthermore, children's acceptance may be more difficult, mainly among younger patients.This can justify why the number of clinical assays involving children is smaller than other age ranges.
More studies on the immune response and the intestinal microbiota during use of probiotics will help to understand AD pathogenesis and to elucidate the controversial results obtained in different studies, thereby contributing to alternative therapies.

Table 2 -
Characteristics of patients after recruitment.Characteristics of patients at baseline.