Atopic dermatitis belongs to the category of atopic diseases and has a prevalence of 10–20 %, one of the most frequent primary manifestation of atopy in children (10–20 %). Atopy is a chronic or chronically recurrent inflammatory skin disease, with concomitant severe pruritus. Children, whose both parents suffer from atopic eczema, have a risk of 60–80 % of developing the disease themselves. Polygenic inheritance is assumed, in which genomic imprinting and various environmental factors also seem to play a role(Reference Burgdorf, Plewig and Wolff1).
The prevalence of atopic diseases, and especially atopic eczema, has increased over the past years(Reference Biedermann and Piche2). There are various hypotheses explaining the increasing prevalence of the allergies. One of these hypotheses is the ‘linoleic acid hypothesis’. It claims that a possible explanation lies in the choice of dietary fats as well as the modified composition of the dietary fats in food(Reference Galland3). A further hypothesis is the ‘hygiene hypothesis’, which argues that the missing infections at a critical time point in the development of the immune system increase the risks for later allergic diseases(Reference Cabana, Michael and Wong4). Various other hypotheses also try to explain the increasing prevalence. However, the cause remains unknown.
Therapy consists of expositional prophylaxis and the administration of glucocorticoids, calcineurin inhibitors and cyclosporine A. Additionally, specific immunotherapy can be performed(Reference Werfel5). Furthermore, in order to avoid atopic diseases, it is recommended to breastfeed 6 months after delivery, avoid passive smoking and protect the child from house dust mites(Reference Biedermann and Piche2).
Probiotics are preparations that contain living micro-organisms, i.e. lactic acid bacteria and yeasts. They may be contained within food or as pharmaceuticals. When ingested in sufficient quantities orally, probiotics may have a health-promoting influence in obstruction, diarrhoea, chronic inflammatory bowel syndrome and other diseases(Reference Bernaola Aponte, Bada Mancilla and Carreazo Pariasca6–Reference Haller, Antoine and Bengmark8).
Some clinical trials confirm that the administration of probiotics already during pregnancy and within the first months of life may reduce the risk for atopic dermatitis(Reference Abrahamsson, Jakobsson and Böttcher9, Reference Kalliomäki, Salminen and Poussa10), whereas other studies(Reference Kopp and Salfeld11) could not show this effect. The gastrointestinal tract of healthy fetuses is sterile. Only during delivery and in the time following, the mother's bacteria colonise the intestine of the fetus and develop into a complex microflora. If probiotics, for example, the Lactobacillus rhamnosus strain GG, are taken during pregnancy, they form part of the mother's gut flora and are thus also transferred to the child. In contrast to the mother, where L. rhamnosus strain GG only remains for a short time after the discontinuation of intake, they remain detectable in the child's stool for another 6 months after delivery and the discontinuation of intake(Reference Schultz, Göttl and Young12).
The safety of the intake of probiotics during pregnancy has been well tested, especially for lactobacilli and bifidobacteria. It is considered to be well tolerated and has a low risk of side effects(Reference Borriello, Hammes and Holzapfel13, Reference Dugoua, Machado and Zhu14).
In the present study, we sought to conduct a systematic review of randomised trials involving the use of probiotics given during pregnancy and the incidence of atopic eczema in children.
Materials and methods
The present study is based on a systematic database research for randomised, controlled studies on probiotic administration during pregnancy and the risk of atopic eczema within the first years of life.
The following databases were searched starting from the respective start of the database up to and including 23 June 2009. The search terms were ‘pregnancy and probiotics’:
(1) PubMed
(2) Ovid
(a) EBM Reviews – Cochrane Central Register of Controlled Trials
(b) EBM Reviews – Cochrane Database of Systematic Reviews
(c) EBM Reviews – Cochrane Methodology Register
(d) EMBASE 1980 until 23 June 2009
(e) Ovid Medline(r) 1950 until 23 June 2009
Subsequently, the references in the publications were searched for additional, potentially important, publications (Fig. 1). Only publications with ethics approval were included.
Fig. 1 Study selection.
Data collection was performed by two independent reviewers while adhering to a data collection sheet. The analyses were then compared and possible discrepancies were solved with the help of a third reviewer (Table 1).
Table 1 Major contents of the studies on probiotics
On the basis of the data collection sheets as well as the original articles, quality assessment was made (Table 2). This was done according to the ‘CRD's guidance for undertaking reviews in health care’ written by the Centre for Reviews and Dissemination. Data that were not found in the original publications could not be considered in the evaluation. An overview of the individual study results is provided in Table 3.
Table 2 Summary of quality criteria
Table 3 Mentioned frequencies, OR, CI and P values in the studies
RR, risk ratio.
The available data were compared with the statistical software Stata/SE 11.0 (StataCorp LP, College Station, TX, USA). It calculated the risk ratio for each study endpoint as well as the respective 95 % CI. In addition, the studies were rated according to their size in order to calculate the influence of individual studies on the meta-analysis. With heterogeneity testing, the comparability of the data that were analysed was assessed.
Results
A total of seven systematic randomised, double-blind and placebo-controlled studies observing 2843 children whose mothers took probiotics or placebo during pregnancy and lactation were included in the meta-analysis. All studies that were included used atopic eczema as an endpoint.
Of those studies, four only used lactobacilli as probiotics, three used a mixture of various bacterial strains (including lactobacilli) and one included bifidobacteria.
On the basis of the selected studies, two meta-analyses were performed. It was observed that one study used lactobacilli and the other studies used a mixture of bacterial strains.
The meta-analysis on those studies that used a mixture of various bacterial strains shows no significant association between the intake during pregnancy and lactation and the development of atopic eczema in the children (P = 0·204). The study by Kuitunen et al. (Reference Kuitunen, Kukkonen and Juntunen-Backmann15) showed the strongest contribution to the meta-analysis (Fig. 2).
Fig. 2 Endpoint analysis of studies that used mixed probiotics and the development of atopic eczema. RR, risk ratio.
The meta-analysis on the studies that used only lactobacilli as probiotics shows a significant correlation between the administration of the probiotics during pregnancy and lactation and the development of atopic eczema (P = 0·045, Fig. 3). All the studies that are included contribute equally to the meta-analysis.
Fig. 3 Endpoint analysis of studies that used lactobacilli and the development of atopic eczema. RR, risk ratio.
Discussion
Overall, probiotics significantly reduce the risk of the development of atopic eczema (P = 0·022). However, the effect can only be ascribed to the results of three of the seven studies.
In a separate analysis of the studies that used lactobacilli and those that used a bacterial strain mixture, only monotherapy resulted in a significant risk reduction for atopic eczema (P = 0·045 v. P = 0·204). Surprisingly, the bacterial load per bacterial strain is comparable in the strain mixture and monotherapy. However, it may be possible that the orally applied bacteria remain in the gut for only a short time due to displacement effects. A possible reason could be a repression of each other, which anticipates the attainment of effective concentrations.
There is some evidence that probiotics maintain the integrity of the intestinal barrier. Some of the effects appear to be mediated through Toll-like receptors, which are also expressed by the enterocytes(Reference Rachmilewitz, Katakura and Karmeli20). But this effect is limited only to some species (Lactobacillus reuteri and Lactobacillus casei) and not to others (Lactobacillus planarum). The reason for the different effects might be that those species cannot bind the three grabbing non-integrin molecules that are blocking the antibodies that are responsible for intercellular adhesion molecule. On the basis of available data, the recommendation for the administration of probiotics consisting of lactobacilli during pregnancy and lactation can be made, as it may lead to a reduction in the development of atopic eczema in children at risk. More longitudinal studies observing the clinical and experimental factors as well as the time of the beginning such a therapy are necessary(Reference Prescott and Bjorksten21).
This effect could not be found in the actual S-3 guidelines from the German Society of Dermatology since two of the publications cited were published after the literature research in March 2008(Reference Schäfer22).
Due to non-significant results, no recommendation for probiotics consisting of different bacterial strains can be given. No evidence-based studies are currently available on other probiotics.
The severity of atopic eczema was less in the group that received L. rhamnosus than in the group that took Bifidobacterium animalis spp. lactis(Reference Wickens, Black and Stanley16).
In conclusion, probiotics, especially lactobacilli, reduce when taken as a monotherapy during pregnancy the child's risk of developing atopic eczema. The long-term development of this effect will have to be assessed in further studies, and so do the possibly differing effects of single bacterial strains.
Acknowledgements
The authors declare no conflict of interest. The study was not funded. The contribution of each author to the manuscript is as follows: K. D. planned the review protocol and performed the review process; D. G. planned the review protocol and advised on the review process; B.-C. Z. planned the review protocol regarding the nutritional facts and advised the manuscript on nutritional facts; E. D. performed the review process; C. z. E. advised on the statistics; K. J. B. advised on all aspects of planning, performing and appraising of the study. All authors approved the final version.
