Probiotic Bacteria in Dietetic Products for Infants: A Commentary by the ESPGHAN Committee on Nutrition

Abstract

INTRODUCTION There is considerable interest in including probiotics in dietetic products for children, and such products are marketed in many countries. This commentary by the ESPGHAN Committee on Nutrition reviews available information on the effects of adding probiotic bacteria to infant formulas, follow-on formulas, and special medical foods. It also discusses the safety of these products and the appropriate conditions for their use. The Committee reviewed expert consensus documents on probiotics in foods and dietetic products for infants. The Committee also systematically reviewed all randomized clinical trials on dietetic products containing probiotics which involved infants. Because probiotics are available in many forms, such as capsules and powders, the Committee reviewed controlled clinical trials that examined the effects of different probiotic preparations for infants. The conclusions of this commentary may require revision in the future as new information becomes available. GUT MICROFLORA Most studies report that the stool flora of breast-fed infants differs from that of formula-fed infants (1). Breast-fed infant stools contain predominantly Bifidobacterium and Lactobacillus, which may account for as much as 90% of the total flora (2). In contrast, the flora of formula-fed infants is more diverse, containing Bacteroides, Bifidobacterium, Staphylococcus, Escherichia coli, and Clostridia (2–4). The species of Bifidobacterium in the stools of breast- and formula-fed infants differ (5). A variety of factors has been proposed as causes for the different fecal flora of breast- and formula-fed infants, including the lower content and different composition of proteins in human milk, its lower phosphorus content, the large variety of oligosaccharides in human milk, and numerous humoral and cellular mediators of immunologic function in breast milk (6). The gut flora appear to modulate health and well-being of the host (7,8). The lower incidence of gastrointestinal and other infections in breast-fed infants (9–11) may in part be related to their gut flora. Bifidobacterium and Lactobacillus inhibit the growth of pathogenic microorganisms through the production of lactic, acetic, and other organic acids, with a consequent decrease of intraluminal pH, whereas formula feeding favors propionate and butyrate production and a near neutral fecal pH. Moreover, Bifidobacterium and Lactobacillus compete with potentially pathogenic bacteria for nutrients and epithelial adhesion sites. The gut flora also promote the recovery of energy and nutrients through fermentation of nondigestible carbohydrate, nitrogen salvage, and beneficial effects on mucosal growth and water and nutrient absorption (12). Evidence is accumulating that gut flora also modulate mucosal physiology, barrier function, and systemic immunologic and inflammatory responses (13). The growing interest in the role of the bacterial gut flora on health has stimulated different strategies to modify the human intestinal flora, including the provision of bacteria considered probiotics and nondigestible carbohydrates considered prebiotics. DEFINITIONS The term probiotic was introduced in 1965 by Still-well and Lilly (14). A widely accepted definition of probiotics is “live microbial food ingredients that are beneficial to health” (15). However, the scientific basis of this definition has recently been questioned because animal studies suggest that some probiotic effects can be achieved by nonviable bacteria and even by bacterial DNA (16–18). Therefore, probiotics have more recently been defined as “microbial cell preparations or components of microbial cells with a beneficial effect on the health and well being of the host” (19). Prebiotics are “non-digestible food components that benefit the host by selectively stimulating the growth and/or activity of one or a limited number of bacteria in the colon, and thereby improve host health” (15). Synbiotics are “mixtures of probiotics and prebiotics that benefit the host by improving the survival and implantation of live microbial dietary supplements in the gastrointestinal tract” (15). EXPERT COMMITTEE REPORTS ON THE USE OF PROBIOTICS IN HUMANS The joint FAO/WHO Expert Consultation on Evaluation of Health and Nutritional Properties of Probiotics in Food Including Powder Milk with Live Lactic Acid Bacteria concluded that probiotic strains belong primarily to two genera, Lactobacillus and Bifidobacterium, which must survive the passage through the digestive tract and proliferate in the large bowel. Enterococcus strains should not be used as probiotic microorganisms (20). It was recommended that strains be named according to the International Code of Nomenclature, be deposited in an internationally recognized culture collection, and that strain identification be performed by phenotypic tests followed by genetic identification with methods such as DNA/DNA hybridization and 16sRNA sequencing (21). Stock cultures should be maintained under appropriate conditions and be checked periodically for strain identity and probiotic properties. Because no in vitro tests predict the probiotic activity of a strain, activity should be determined in clinical trials following accepted standards of scientific quality. Beneficial effects must be related to dosage regimens and duration of use in each product or strain. Safety considerations should include transmission of antibiotic or drug resistance inherent in some probiotic microorganisms. The Expert Consultation further recommended that dried milk powders containing live lactic acid bacteria should preserve adequate numbers of viable probiotic bacteria with stable probiotic properties throughout shelf-life, and that labeling should include the microbial species or strain and the proportion of viable organisms. Claims for health benefits should be substantiated with scientific evidence. Subsequently published Guidelines for the Evaluation of Probiotics (22) further emphasize the need to fully evaluate the safety of probiotics, in particular the risk of infection in subjects with compromised immunity and subjects at risk for endocarditis. The French Agency for Food Safety (AFFSA) reviewed the safety of probiotics in infants, excluding genetically modified and nonviable microorganisms from its definition of probiotics (23). The report distinguished two distinct periods of modification of intestinal colonization: the first week of life, when breast-fed infants develop a flora dominated by bifidobacteria and formula-fed infants develop a more complex flora, and the period when complementary feeding is started. The report concluded that the possible health consequences of the enzymatic activities of the more diversified flora of formula-fed infants compared with that of breast-fed infants are not known. For infants without lactose intolerance, the lactose-fermenting capacity of probiotic bacteria confers no appreciable benefit. No conclusion was made as to whether reduced bacterial translocation, seen in animals supplemented with certain probiotic strains, occurred in infants. The Agency recommended for safety reasons that probiotics should not be given to immunocompromised or premature infants. In addition to requirements on strain identity, viability, strain stability, and number of probiotic bacteria in a food at the end of its shelf life, the Agency recommended that instructions for preparation, storage, and heating of the formula be specific to guarantee the survival of the desired number of microorganisms until the time of feeding. The Agency further recommended that nutritional, physiological, and therapeutic effects be documented in appropriate clinical studies. The Scientific Committee on Food of the European Commission also commented on the use of probiotic bacteria in food products for infants (24). It recommended that infant formulas with probiotic microorganisms should be marketed only if their benefit and safety have been evaluated according to the principles outlined by the same Committee. The Committee did not object to the addition of probiotic bacteria to follow-on formulas. However, the Committee stated that only bacterial strains with identity and genetic stability demonstrated by cultural and molecular methods should be used. The identity of the probiotic strain should be described by molecular methods in a dossier and be available to the food control authorities. The content of viable bacteria should be adequate throughout shelf-life to achieve 106 to 108 colony-forming units (CFU) per gram of formula prepared as ready for consumption. SYSTEMATIC REVIEW OF CLINICAL TRIALS ON DIETETIC PRODUCTS WITH PROBIOTICS IN INFANTS Three databases (MEDLINE, EMBASE, and Cochrane Controlled Trials Register) were searched up to July 2003. We reviewed randomized and quasirandomized (i.e., allocating participants according to date of birth, the number of hospital records, etc.) controlled trials (RCT) of infant or follow-up formulas, or special medical foods, supplemented with bacteria generally considered probiotics. All references to review articles in the identified trials were reviewed. A separate search was made using the names of authors considered experts in this field. No limit was imposed as to the language of publication. Letters to the editor, abstracts, and proceedings from scientific meetings were excluded. Only clinically important outcomes (end points) were considered. Surrogate outcome measures (laboratory parameters) were excluded. After the exclusion criteria were applied, the search strategies yielded six articles on the clinical effects of feeding dietetic products supplemented with probiotics, including four on infant or follow-on formulas (25–28) and two on special medical foods (29,30) (Table 1). When scoring trial quality using Jadad’s criteria (31), only one study was of good methodological quality (25).TABLE 1: Trials with dietetic products for infants and young children supplemented with probioticsTABLE 1: ContinuedA. Clinical Effects Infant and follow-on formulas. One high-quality double-blind RCT of 55 infants and young children aged 5 to 24 months in a chronic medical care hospital in a developed country found that administration of standard infant formula supplemented with Bifidobacterium lactis (formerly called Bifidobacterium bifidum) (1.9 × 108 CFU/g powdered formula) and Streptococcus thermophilus (0.14 × 108 CFU/g) reduced the prevalence of nosocomial diarrhea compared with placebo (7% versus 31%; relative risk [RR]: 0.2; 95% confidence interval [CI]: 0.06–0.8). The risk of rotavirus gastroenteritis was significantly lower in those receiving probiotics supplemented formula (RR: 0.3; 95% CI: 0.09–0.8). Feeding B. bifidum and S. thermophilus led to a significantly lower rate of rotavirus shedding (25). One low quality RCT (no blinding, allocation concealment unclear, dropout and withdrawal rate not documented) involving 175 children living in an orphanage in a developing country reported no protective effect of infant formula supplemented with Bifidobacterium Bb12 (108 CFU/g) alone on episodes of diarrhea (40/62; 65%) or in combination with S. thermophilus (dose not given) (29/56; 52%) compared with placebo (14/57; 25%). Rotavirus accounted for only 3 of 81 (3.7%) episodes of diarrhea, and bacterial pathogens accounted for 45 of 81 (56%) episodes of diarrhea (26). Foods for special medical purposes. Two small RCTs of infants with atopic dermatitis and cow’s milk allergy during formula feeding or breastfeeding were found (29,30). Neither study provided details of randomization and blinding. Allocation concealment was unclear. In the first study (29), infants (n = 27) with atopic eczema and cow’s milk allergy were randomly assigned to receive extensively hydrolyzed whey formula supplemented with Lactobacillus GG (5 × 108 CFU/g formula) or placebo for 1 month. There was a statistically significant reduction in the clinical score of atopic dermatitis (SCORAD) during the 1-month study. However, by 2 months the SCORAD was similar in both groups. In the second study (30), infants (n = 27) with atopic eczema during exclusive breastfeeding were randomly assigned to receive extensively hydrolyzed whey formula supplemented with Lactobacillus GG (3 × 108 CFU/g), or Bifidobacterium lactis Bb-12 (1 × 109 CFU/g), or the same formula without probiotics. No details on the duration of intervention were given. After 2 months, a statistically significant reduction in SCORAD score was observed in the groups consuming probiotics compared to placebo. A significant change in the SCORAD scores at the 2-month evaluation was seen in 9 of 9 patients receiving B. lactis Bb-12, and in 9 of 9 patients in the Lactobacillus GG group, as compared with 4 of 9 patients not receiving probiotics. In conclusion, there are very limited published data on the clinical effects of probiotic supplementation of infant formulas, follow-up formulas, and special medical foods. Although some short-term benefits are scientifically demonstrable, until more studies are available it is not possible to conclude that the clinical effects of probiotic supplementation are or therapeutic for B. It is recommended that the safety of breast milk be evaluated are introduced to the important part of this evaluation is the on have growth data from studies the effect of infant formulas and foods for special medical purposes. Infant and follow-on formulas. Three RCTs growth of children formulas supplemented with probiotic bacteria were found (Table The first RCT evaluated infants randomized to formula with or without Streptococcus thermophilus and Lactobacillus for the first 2 months of There were infants in each group, and the two groups were compared with a of fully breast-fed infants. and were at birth, 1 and 2 No data on growth were given. It was stated that that the of feeding growth during the first 2 months of without significant there was no information on this conclusion was In the second RCT infants aged 5 to 24 months with chronic and in a chronic medical care hospital were randomized to receive standard infant formula (n = or the same formula supplemented with Bifidobacterium bifidum and Streptococcus thermophilus (n = The duration of was 81 There was no in the and of the groups at was no data were on growth during the and the groups were not It was stated only that all infants maintained or their during the study. The RCT of children aged months was in a in The children were randomized to formulas, which for one infant formula with Bifidobacterium lactis Bb-12 and Streptococcus thermophilus (n = one the same formula with only Bb-12 (n = and one formula without probiotics (n = and were and the change in standard score was used to evaluate There were no significant in rate of the groups. The two groups with probiotics a significantly growth the end of the intervention did the control The was to the is of quality because no details of randomization were allocation concealment was unclear, there were significant the study groups in of no data were given on of formula and the follow-up at months was (Table 1). Foods for special medical purposes. Only one study has growth in infants randomly assigned extensively hydrolyzed whey formula with Bifidobacterium lactis Bb-12, Lactobacillus or no probiotics and were at and 2 and months of It was stated that the growth of all children was the groups were not In conclusion, only one study has provided growth data adequate to the effects of infant formula with probiotics. This study was performed in infants living in a in One of that study was its large from to months, because growth and of growth are different in young infants and This study found a effect of probiotics on because the is of the conclusion is of limited The other studies were small with to effects on of the studies did not the growth a of no conclusions can be made the of probiotic bacteria in dietetic products on infant However, there are no from the available data that probiotics have effects on CLINICAL TRIALS ON THE USE OF PROBIOTICS IN There is on the effects of probiotics in other powdered documented are the therapeutic effects of certain probiotic strains in as recently in two reviews A beneficial effect was found in the of diarrhea by which was strain and benefit also to be achieved in diarrhea by other no was found in bacterial Beneficial effects were more when with probiotics was in the of Two reviews that probiotics given in combination with the risk of diarrhea In the and strains of have been used in this However, with the trials were performed in and the conclusion may not be to the Committee at the studies in trials were One small RCT children with and found no significant of diarrhea 95% CI: A further double-blind RCT at and Bifidobacterium methodological (i.e., small of definition of diarrhea as an end conclusions Two RCTs a effect of GG for diarrhea in children in are dosage of Lactobacillus GG used in in and in the of diarrhea may be for There are no randomized studies in children on the incidence of diarrhea by Three studies of the of probiotics for nosocomial diarrhea were One small trial beneficial effects of Bifidobacterium bifidum and Streptococcus thermophilus (25). Evidence for the of Lactobacillus GG is with one RCT benefit and different A diarrhea was in a trial of Lactobacillus GG used in a developing country in a with a of This benefit was in children to months of were not breast = However, no effect was found in a similar trial in In a study of children 1 to of from care milk containing Lactobacillus GG no significant diarrhea, as by the number of with or the proportion of children without diarrhea during the study However, the with Lactobacillus GG to have with a reduction in the number of by gastrointestinal and during the study = The same study demonstrated a reduction of infections (RR: 95% CI: number to 95% CI: and antibiotic (RR: 95% CI: 95% CI: in children milk with Lactobacillus After for of these was statistically In the same significant reduction in the risk of was seen = controlled for and = the use of Lactobacillus GG is more has not been study from found that Lactobacillus GG (1 × given to the for 2 to 4 and to the for months reduced the incidence of atopic eczema in children at 2 (RR: and 4 of (RR: One study that a combination of and given for to children 1 to of be beneficial in the of atopic dermatitis of possible and of the clinical effects by further in other is from animal and patients with inflammatory have the of the in the of the in clinical probiotics have only limited In children with a trial of Lactobacillus GG that Lactobacillus GG reduced intestinal and activity However, a randomized study children with small and large in that Lactobacillus GG in a of 109 was not in when to standard for In children, conclude that the health effect of probiotics is the reduction of the duration of The benefit of probiotics in other conditions is to be However, there is a of to suggest that may be in the of nosocomial and diarrhea, and Effects appear to be strain specific and be from strain to strain. REPORTS ON THE USE OF PROBIOTICS IN INFANTS and low are risk factors for and from and It has been that administration of probiotics to premature with of intestinal different from those of infection and and the use of studies have examined the of probiotics to the gut of and infants. on the of Lactobacillus GG have rate was reported in a randomized controlled trial of One RCT in very infants found that infants were supplemented with Bifidobacterium of fecal at 2 of and feeding Clinical outcomes were not reported In the administration of and Bifidobacterium to all in an care with a incidence of reduced the incidence of compared with The study is limited by not being A randomized controlled trial in infants found that Lactobacillus GG supplementation × 109 with the first feeding did not tract or In only a limited number of controlled trials have health outcomes following administration of probiotic in infants. studies are The safety of probiotics has been extensively reviewed recently Although the probiotics used in clinical trials generally have been described as and well there have been some that this conclusion In microorganisms in probiotics have been from patients with or infections and infections with Lactobacillus species in infants and children have been reported all patients with infections in these studies have conditions to in the of or in the of In of the to have from the Only a limited number of have been reported in which the was not to be related to the use or of a probiotic product In these patients conditions were of infection with Bifidobacterium during supplementation with this have not been A report from that the use of Lactobacillus GG in food has not in an incidence of Lactobacillus or in the proportion of Lactobacillus all of with probiotics has not been reported in infants or children is a of the published have concluded that the risk of infection with probiotic or bifidobacteria is similar to that of infection with strains, and that of such products is a risk to including immunocompromised effects in which probiotics a role include and However, the available data from and clinical not that such effects with the probiotic in use. In probiotics used in clinical trials can be generally considered as However, for possible such as infection in is and is review of available clinical trials found only limited data on the safety and clinical effects probiotic preparations to infant formulas, follow-up formulas, and special medical foods. There is no published for clinical benefit of infant formulas supplemented with probiotic No data are available on possible effects on intestinal and its effects on gastrointestinal and of such data be given the that bacteria during are more to the those during There are some data a short-term benefit of some probiotic strains in infants and young children with The Committee that when adding probiotics to dietetic products for infants, only bacterial strains for which identity and genetic stability have been demonstrated by cultural and molecular methods and strains considered as generally when to the food in should be used. The content of viable bacteria in dietetic products must a to be and with to defined outcomes in clinical trials throughout the shelf-life of the The Committee that further evaluation of the safety and of probiotic bacteria in dietetic products for infants is strain to be used must be evaluated at the of for and and must be safety that should be are the effects on nutrient the possible of antibiotic the and effects on the and the risk of The Committee is that the available data are not to the safety of probiotics in and very young infants with infants with immunocompromised premature infants, and infants with The Committee that infant formulas with bacteria as probiotics should be marketed only if a evaluation of benefits and safety following the principles of standards has been Although the available data on the use of supplemented follow-on formulas is the Committee has effects of these products because are for use in infants 5 months, when there is a more an intestinal and a of to a variety of from the The addition of probiotic bacterial strains to infant foods for special medical and used under medical may be if a clinical benefit has been in adequate clinical even if a evaluation of all safety has not been for because of the limited number of patients to be with a the specific dietetic The Committee that there is that some probiotic preparations have benefits on health and benefits include a reduced of diarrhea, effects on diarrhea, of in vitro and animal studies on digestive and and from human studies on possible short-term and therapeutic effects on atopic In of the for benefits on health that be achieved by the use of some probiotic on their evaluation are