Effect of synbiotic supplementation on children with atopic dermatitis: an observational prospective study

Atopic dermatitis (AD) is a common chronic, relapsing inflammatory allergic disease with highly pruritic skin lesions, particularly prevalent in children. It usually starts in the first 5 years of life and often has a profound negative effect on the quality of life of patients and their families [1]. AD affects up to 20% of children and 3% of adults, with the prevalence increasing globally [18].

The successful management of AD includes hydration, restoration of the skin barrier, control of skin inflammation, and treatment of secondary infections [10, 14]. Topical corticosteroids remain the first-line medical treatment for the control of symptoms, but relapses are common [20] and adverse effects limit their chronic use [11]. Calcineurin inhibitors are sometimes effective in reducing inflammation and help spare the use of topical steroids [5].

Synbiotics have been defined as combinations of pre- and probiotics with a synergistic action on human health [8]. Prebiotics are food components that induce the growth or activity of probiotics, which are living organisms that, when administered in adequate amounts (at least 10⁹ colony-forming units [CFU]), can be beneficial in the treatment of various conditions. Widely known probiotics such as Bifidobacteria and Lactobacilli have been identified as key components for proper immune system stimulation and homeostasis of the gastrointestinal tract microenvironment [7, 21]. Dysbiosis with increased levels of Clostridium and low levels of Bifidobacterium species in the intestinal microbiota has been observed in atopic children and was speculated to contribute to inflammation in AD [13]. Balancing the gut microflora may improve gut barrier function and reduce the production of proinflammatory cytokines. In addition, several in vitro and clinical studies showed that consumption of probiotics suppressed Th2 response and shifted Th1/Th2 balance towards Th1 response [16, 17]. Recently, modulation of microbiota to promote clinical improvement in pediatric patients with AD has been a focus of increasing interest. Meta-analyses of randomized controlled trials have shown that multistrain pro- and synbiotics are of benefit for the prevention and treatment of AD in children [2‐4, 6].

The main objective of this prospective observational study was to assess the effect of a multistrain synbiotic supplement containing Lactobacillus casei, Bifidobacterium lactis, Lactobacillus rhamnosus, and Lactobacillus plantarum plus oligosaccharides and biotin for 8 weeks in children with AD. Treatment tolerance was also evaluated.

The rate of recruitment was lower than expected, and a total of 353 patients were recruited within the foreseen period. Thirty-three (9.3%) patients did not meet the selection criteria and were excluded. Therefore, the study population included 320 patients, of which 275 completed the study. The flow diagram of participants is shown in Fig. 1.

Baseline characteristics are summarized in Table 1. Most (54.9%) patients were girls, and the median age was 4 years (range 3 months to 11.8 years). Concomitant respiratory allergy and food allergy were reported in 36.9% and 21.3% of patients, respectively; history of parental atopy was present in 70.9% of cases, most frequently AD or respiratory allergy. The median duration of AD was 3.4 years (IQR 1.7–6.2 years). Persistent AD was present in 52.1% of patients. Less than half of patients (43.4%) were receiving concomitant treatments, most frequently topical corticosteroids (21.3%), antihistamines (19.7%), or treatments for concomitant diseases (“Others,” Table 1). Thirty-three (11.1%) patients had changes in their concomitant treatment between the baseline and the follow-up visits.

At baseline, the mean (SD) SCORAD was 45.5 (15.5) (95% CI 43.8–47.3) (Table 2), and more than half of the patients (57.3%) had moderate disease (SCORAD 25–50) (Fig. 2). The intensity of the cutaneous symptoms was moderate in most patients. The mean VAS score for pruritus was 5.7 (2.2) (95% CI 5.5–5.9) and for sleep loss 3.1 (2.5) (95% CI 2.9–3.4). At the final visit (week 8), the mean SCORAD was 19.4 (14.6) (95% CI 17.7–21.2), cutaneous lesions had cleared or were of mild intensity in most patients, and the mean VAS score for pruritus was 2.3 (2.2) (95% CI 2.0–2.5) and for sleep loss 1.1 (1.8) (95% CI 0.9–1.3). Average (SD) intra-patient difference in SCORAD score between the basal and the follow-up visits was 27.0 (15.1) points (median 27.0; range − 6 to 79.3). All differences in SCORAD were statistically significant (P < 0.001) (Table 2). The number of patients with severe disease decreased from 35.1% at baseline to 4.4% at week 8 (Fig. 2). In the PP dataset, the mean SCORAD index decreased from 47.3 (15.5) (95% CI 45.3–49.3) at baseline to 19.2 (14.8) (95% CI 17.3–21.1) after 8 weeks of treatment (P < 0.001).

The percentage of physicians rating AD as “very much improved,” “much improved,” “no change,” and “worse” was 35.9%, 49.8%, 13.9%, and 0.3%, respectively. These percentages were similar to the parents’ opinion (38.9%, 42.7%, 17.4%, and 1.0%, respectively) (kappa 0.673). The kappa index for concordance between improvement/no improvement and 30% and 50% improvement of the SCORAD was 0.698 and 0.443 for physicians, and 0.643 and 0.520 for the parents.

In multiple linear regression analysis, the independent predictors of stronger improvement (that is, a greater decrease in the SCORAD index between baseline visit and week 8) were higher baseline SCORAD (OR 0.51, 95% CI 0.41–0.61; P = 0.0001) and adherence ≥ 80% (OR 7.29, 95% CI 1.85–12.73; P = 0.009) (Table 3).

A total of 29 adverse events were recorded in 21 patients (Table 4). Of them, only two events of abdominal pain observed in the same patient were judged by the investigator to be potentially related to the product.

This study carried out in real-world practice shows that an oral supplementation with a synbiotic product with high levels of viable organisms for 8 weeks was effective in improving AD in children. The product included a mixture of four bacterial strains of commensal organisms (L. casei, L. rhamnosus, L. plantarum, and B. lactis), fructooligosaccharide, galactooligosaccharide, and biotin, the benefits of which in several immune-mediated and allergic diseases has been documented [15, 19, 23, 25]. Importantly, improvement was observed in long-standing AD (mean time from diagnosis of 4 years), both in persistent disease and in AD with flares, in patients with and without concomitant atopies, as well as in the presence or absence of underlying treatment for AD. These variables did not affect improvement of SCORAD index in bivariate analyses. Treatment was well tolerated, and of the 29 registered adverse events, only two events of abdominal pain were judged to be possibly related to the synbiotic product.

We used the SCORAD index, a valid and reliable tool, for the measurement of the main outcome of the study [22]. Statistically significant improvements in the total score as well as in the intensity of six cutaneous symptoms, pruritus, and sleepiness were found after 8 weeks of treatment. Also, the level of agreement of physicians and parents regarding amelioration of AD was substantial for a 30% decrease of the SCORAD score and moderate for a 50% decrease.

Conflicting data on the effect of probiotics on AD have been reported in literature, although in general, results favor probiotics over controls. A meta-analysis of 25 RCTs with 1599 AD patients found a greater improvement in SCORAD in the probiotics group compared to controls in children 1 to 18 years old (difference in change in SCORAD − 5.74 points vs control, 95% confidence interval − 7.27 to − 4.20), and in adults (difference in change in SCORAD − 8.26 points vs control, 95% confidence interval − 13.28 to − 3.25), although the effect was not proven in infants (< 1 year old) [12]. Similarly, a meta-analysis on RCT on synbiotics in children with AD (6 studies, 369 children) found a greater decrease in SCORAD (by 6.56 points) in synbiotic groups vs placebo (95% CI, − 11.43 to − 1.68) [6]. Heterogeneity was high in both meta-analyses, and in both cases, the authors found greater benefits in case of treatment with mixed bacterial species [6, 12].

Yesilova et al. [24] carried out a study in 40 children with AD who were randomized to supplementation with a probiotic mixture or placebo. Similarly, these patients were treated for 8 weeks with a high-dose (2 × 10⁹ CFU) multistrain probiotic complex containing B. bifidum, L. acidophilus, L. casei, and L. salivarius. The authors observed a significant decrease of SCORAD values at the end of treatment, with the magnitude of SCORAD reduction (mean change from baseline 23.0 points, compared to 12.8 points in placebo arm) comparable to that in our study (mean change from baseline 26.1 points). The probiotic supplement was also effective for reducing serum IL-5, IL-6, IFN-γ, and total serum IgE levels; unfortunately, the effect on IgE levels and proinflammatory cytokines was not examined in our study.

The improvement in SCORAD index and in all symptoms that was observed in our study was stronger than in most published controlled trials on probiotics in AD, which could be due to various reasons, including differences in study design, dose and strains of probiotics used, duration of intervention, characteristics of the study population, and the sample size. Since this was a single-arm, non-controlled study, we cannot rule out the placebo effect and/or improvement due to the natural course of the disease. It should be noted, however, that many participants had a long-standing AD (mean time since diagnosis 4 years), and about half had persistent disease, so it seems unlikely that all the improvement was due to the natural progression of the disease. Besides, since this was an observational study, there were no restrictions on concomitant medication, and 21% of study participants were receiving topical corticosteroids and 19% antihistamines. Overall, however, the percentage of patients receiving medical treatment for AD was lower than what could be expected in this population (most patients with moderate or even severe disease), which suggests that AD in our practice may be undertreated. Importantly, patients with and without concomitant medication showed a similar improvement in AD, suggesting that the observed changes could not be attributed exclusively to concomitant treatments. Furthermore, we found a significant association between the adherence to treatment with the multistrain high-dose synbiotic product and decrease of SCORAD index. Good adherence was the most important predictive factor of clinical improvement both in the bivariate and multivariate analyses, which further supports the idea that at least part of the observed improvement in AD was due to the synbiotic treatment.

One possible limitation affecting the validity of the results was that the participant rate was lower than the projected sample size. The observational character of the study allowed including a varied sample of patients, and the multivariate analysis allowed assessing the effectiveness of treatment in relation to patient’s characteristics. Well-designed RCTs should be conducted to elucidate the effectiveness of synbiotics in AD treatment. Including more representative real-world patient samples and complementing RCT with data from observational studies should help physicians treating children with AD to have appropriate evidence on which to base their clinical decisions in daily practice.

In conclusion, our results indicate that supplementation with multistrain high-dose synbiotics (L. casei, L. rhamnosus, L. plantarum, and B. lactis, combined with fructooligosaccharides, galactooligosaccharides, and biotin) improves AD in children, whereas the tolerability and safety profile are very good. Further studies, including RTCs, are needed to add evidence on the benefits of synbiotics in the treatment of pediatric AD in real-life clinical practice.