Probiotics in critically ill children

Gut microflora contribute greatly to immune and nutritive functions and act as a physical barrier against pathogenic organisms across the gut mucosa. Critical illness disrupts the balance between host and gut microflora, facilitating colonization, overgrowth, and translocation of pathogens and microbial products across intestinal mucosal barrier and causing systemic inflammatory response syndrome and sepsis. Commonly used probiotics, which have been developed from organisms that form gut microbiota, singly or in combination, can restore gut microflora and offer the benefits similar to those offered by normal gut flora, namely immune enhancement, improved barrier function of the gastrointestinal tract (GIT), and prevention of bacterial translocation. Enteral supplementation of probiotic strains containing either Lactobacillus alone or in combination with Bifidobacterium reduced the incidence and severity of necrotizing enterocolitis and all-cause mortality in preterm infants. Orally administered Lactobacillus casei subspecies rhamnosus, Lactobacillus reuteri, and Lactobacillus rhamnosus were effective in the prevention of late-onset sepsis and GIT colonization by Candida in preterm very low birth weight infants. In critically ill children, probiotics are effective in the prevention and treatment of antibiotic-associated diarrhea. Oral administration of a mix of probiotics for 1 week to children on broad-spectrum antibiotics in a pediatric intensive care unit decreased GIT colonization by Candida, led to a 50% reduction in candiduria, and showed a trend toward decreased incidence of candidemia. However, routine use of probiotics cannot be supported on the basis of current scientific evidence. Safety of probiotics is also a concern; rarely, probiotics may cause bacteremia, fungemia, and sepsis in immunocompromised critically ill children. More studies are needed to answer questions on the effectiveness of a mix versus single-strain probiotics, optimum dosage regimens and duration of treatment, cost effectiveness, and risk-benefit potential for the prevention and treatment of various critical illnesses.


Introduction
Critically ill patients are predisposed to altered gut microflora, which can lead to infective and non-infective complications and adverse outcome 1-3 . Probiotic bacteria have the potential to restore the balance of gut microflora in critically ill children and confer a health benefit when given for various indications. Probiotics are defined by a joint working group of the Food and Agriculture Organization of the United Nations/World Health Organization as "live microbes which when administered in adequate amount confer health benefit to the host" 4 . In addition, probiotics should be non-pathogenic, stable in acid and bile, able to adhere to and colonize human gut mucosa, and retain viability during storage and use. They should be scientifically demonstrated to have beneficial physiological effects and safety so that they can be used to improve microbial balance and to confer health benefit. In recent years, probiotics have been increasingly used in critical care settings for the prevention of certain diseases that are otherwise associated with high mortality. In this review, we examine the current status of probiotics in the care of critically ill children on the basis of available literature and identify directions for future research.

Gut microflora
The human gut represents a complex ecosystem where a delicate balance exists between the host and the microflora. More than 400 different species of microbes live in the gut as commensal; the total estimated number is more than 10 times the number of eukaryotic cells in the human body 3,5 . Human gut microflora consists principally of obligate anaerobes (95%; Bifidobacterium, Clostridium, Eubacterium, Fusobacterium, Peptostreptococcus, and Bacteriodes) and facultative anaerobes (1-10%; Lactobacillus, Escherichia coli, Klebsiella, Streptococcus, Staphylococcus, and Bacillus). Bifidobacteria are predominant microbes that represent up to 80% of the cultivable fecal bacteria in infants and 25% in adults. Each human being has his or her own unique microbial composition, especially of lactic acid bacterial (LAB) strains 3 . Most of these microbes have health-promoting effects; however, a few are potentially pathogenic. Normally, the 'good' microbes outnumber potentially pathogenic bacteria and live in symbiosis with the host. The optimal balance, composition, and function of gut microflora depend on the supply of food (fermentable fibers and complex proteins) and fluctuate with antibiotic usage, diarrheal diseases, and critical illness 3 . The gut microflora benefits the host by performing various crucial functions (Table 1).

Critical illness and gut microflora
Critical illness and its treatment create a hostile environment in the gastrointestinal tract (GIT) and alter the microflora that tilts the balance to favor overgrowth of pathogens. The hostile environment is exacerbated by the use of broad-spectrum antibiotics, invasive central lines, endotracheal intubation, mechanical ventilation, antacids, H 2 blockers, steroids, and immunosuppressive and cytotoxic therapy. Multiple organ dysfunction syndrome (MODS), burns, malnutrition, changes in nutrient availability, gut motility, pH, redox state, osmolality, and the release of high amounts of stress hormones (including catecholamines) further compromise the critical balance 2,3 . Studies in experimental models have shown that after onset of acute pancreatitis there was disappearance of beneficial LAB within 6 to 12 hours 6-8 . In patients with systemic inflammatory response syndrome (SIRS), there is a reduction in beneficial bacteria (Bifidobacterium and Lactobacillus) that leads to a decrease in short-chain fatty acid levels and elevation of intestinal pH, indicating a disturbed intestinal environment 9 . Hostile gut environment and disruption of the balance of gut microflora alter local defense mechanisms and lead to colonization and overgrowth of potentially pathogenic commensals such as Salmonella, E. coli, Yersinia, and Pseudomonas aeruginosa. These pathogenic commensals cause cytokine release, cell apoptosis, activation of neutrophils, and disruption in epithelial tight junctions 1,2 . With loss of "colonization resistance", the gut is unable to prevent the translocation of pathogens and toxins across the gut wall into the bloodstream, leading to SIRS, MODS, and mortality. Interestingly, the gut has been identified as the originator and promoter of health care-associated infections (HCAIs) and MODS in critically ill patients 1,10 . Restoring the beneficial gut microflora with an exogenous supply of new and effective microbes (probiotics) seems an attractive option to restore the "colonization resistance".

Commonly used probiotics
The most frequently used probiotic strains are Lactobacillus and Bifidobacterium 11 ; other species of probiotics are enlisted in Table 2. These probiotics are used either singly or in combination. Multistrain probiotics are likely to be better than single-strain probiotics, as individual probiotics have different functions and have synergistic effects when administered together. A daily intake of 10 6 -10 9 colony-forming units (CFUs) is reportedly the minimum effective dose for therapeutic purposes 11,12 .

Beneficial functions Details of beneficial functions
Immune response Gut microflora stimulate the proliferation and differentiation of epithelial cells in large and small intestines, modulate innate and adaptive immune response and development of competent gutassociated immune system, and maintain an immunologically balanced inflammatory response 5,61,62 .
Physical barrier function (colonization resistance) Gut microbiota provide a physical barrier against pathogen invasion by competing for epithelial cell adhesion sites, preventing epithelial invasion, competing for available nutrients affecting the survival of potential pathogens, and producing anti-bacterial substances (e.g. bacteriocins and lactic acid), making the environment unsuitable for the growth of pathogens 3,63 .

Nutritive functions
Gut microbiota produce several enzymes for fermentation of non-digestible dietary residue and endogenously secreted mucus and help in recovering lost energy in the form of short-chain fatty acids 64 . They also help in the absorption of calcium, magnesium, and iron; synthesis of vitamins (folic acid and vitamin B1, B2, B3, B12, and K); biotransformation of bile acids; and conversion of pro-drugs to active metabolites [64][65][66] .

Mechanism of beneficial effects of probiotics
The beneficial effects of probiotics are due to change in the composition of gut flora and modification of immune response 13 . Probiotic strains activate mucosal immunity and stimulate cytokine production, IgA secretion, phagocytosis, and production of substances (such as organic acids, hydrogen peroxide, and bacteriocins) that are inhibitory to pathogens. They also compete for nutrients with pathogenic bacteria and inhibit pathogen attachment and action of microbial toxin. Probiotics also have a trophic effect on intestinal mucosa (by stimulating the proliferation of normal epithelium that maintains mucosal barrier defenses), modulate innate and adaptive immune defense mechanisms via the normalization of altered gut flora, and prevent bacterial translocation 12-16 . Table 3 and Table 4 provide a summary of various studies demonstrating different mechanisms of action of probiotics in experimental and clinical studies, respectively.  Sánchez et al. 70 In rats with carbon tetrachloride-induced cirrhosis VSL#3 Decreased incidence of bacterial translocation in VSL#3 group than in water group (8% versus 50%; P = 0.03)

Probiotic use in critically ill children
Studies have evaluated the role of probiotics in critically ill children for the prevention and treatment of necrotizing enterocolitis (NEC), antibiotic-associated diarrhea (AAD), and HCAIs, including ventilator-associated pneumonia (VAP), Candida colonization, and invasive candidiasis.

Probiotics and necrotizing enterocolitis
In 1999, a study showed that oral administration of Lactobacillus acidophilus and Bifidobacterium infantis reduced NEC 17 . This was followed by a negative study showing that 7 days of L. rhamnosus GG supplementation starting with the first feed was not effective in reducing the incidence of urinary tract infection, NEC, or sepsis in preterm infants 18 . However, subsequent randomized controlled trials (RCTs) with different strains of Lactobacilli and Bifidobacteria showed a significant reduction in the development of NEC 19,20 . A systematic review and meta-analysis by Alfaleh et al. 21 in 2008 concluded that probiotic supplementation reduced the incidence of NEC stage II (or more) and mortality. A more recent meta-analysis by the same authors, involving 24 trials in preterm neonates, found that supplementation with probiotic preparations containing Lactobacillus either alone or in combination with Bifidobacterium prevents severe NEC and reduces all-cause mortality 22 .
Probiotics in antibiotic-associated diarrhea The osmotic and invasive AAD is often observed among critically ill children receiving broad-spectrum antibiotics. It is attributed to overgrowth of pathogens and a decrease in population of microbes that have beneficial metabolic functions 23 . Several investigators have shown that probiotics could prevent AAD. The results of metaanalyses on the effect of probiotics for the prevention of AAD are given in Table 5.  33 , in a study of 249 preterm neonates who were subdivided to receive L. reuteri (n = 83), L. rhamnosus (n = 83), and no supplementation (n = 83), found that both the probiotics were effective in reducing Candida colonization in the GIT, late-onset sepsis, and abnormal neurological outcomes. Another RCT, by Demirel et al. 34 , found that in VLBW infants (gestational age of not more than 32 weeks and birth weight of not more than 1500g) prophylactic Saccharomyces boulardii supplementation was as effective as nystatin in reducing fungal colonization and invasive fungal infection and was more effective in reducing the incidence of clinical sepsis and number of sepsis attacks. An RCT by Roy et al. 35 demonstrated that supplementation with a mix of multiple probiotics (a mix of L. acidophilus, B. longum, Bifidobacterium bifidum, and Bifidobacterium lactis) in preterm infants and neonates led to reduced enteral fungal colonization and invasive fungal sepsis, earlier establishment of full enteral feeds, and reduced duration of hospital stay. More recently, Oncel et al. 36 , in a RCT, demonstrated that prophylactic oral administration of L. reuteri in preterm infants (gestational age of not more than 32 weeks and birth weight of not more than 1500g) was as effective as nystatin in the prevention of fungal colonization and invasive candidiasis and reduced the incidence of sepsis, feeding intolerance, and duration of hospitalization.
Limited data are available on the role of probiotics in the prevention of Candida colonization and Candida infection in critically ill pediatric patients. In a placebo-controlled RCT, we found that administration of a mix of probiotics (L. acidophilus, L. rhamnosus, B. longum, B. bifidum, S. boulardii, and S. thermophilus) for 1 week to children being treated in a PICU with broad-spectrum antibiotics decreased the prevalence of Candida colonization of the GIT by 34.5% and 37.2% on days 7 and 14, respectively, and led to an almost 50% reduction in the incidence of candiduria 37 . We also observed that the rate of Candida bloodstream infection was lower in the probiotic group as compared with the placebo group; the difference, however, was not statistically significant, as the sample size was not sufficient to evaluate this outcome. To test the hypothesis that the enteral supplementation with probiotics in critically ill children can decrease the prevalence of invasive candidiasis, we conducted a retrospective "before and after" study that included critically ill children on broad-spectrum antibiotics for at least 48 hours. The study showed that the probiotics group (4 of 344, 1.2%) had a significantly lower incidence of candidemia than the control group (14 of 376, 3.7%, RR 0.31; 95% CI 0.10-0.94; P = 0.03) 38 . Candiduria was noted in 10.7% of patients in the probiotic group and 22% in the control group (RR 0.48; 95% CI 0.34-0.7; P = 0.0001) 38 .
Complementing these clinical studies, laboratory studies have also shown that several probiotic strains prevent Candida colonization by inhibiting adhesion and biofilm formation, germination, and conversion of yeast to germ (filamentation) 14,39 . Overall, the current evidence shows that supplementation of probiotics could be a potentially effective strategy in reducing Candida colonization as well as invasive candidiasis in critically ill children.

Safety of probiotics
Although most commercially available probiotic strains are widely regarded as safe, there are some concerns with respect to safety, particularly in severely debilitated or immunosuppressed patients 3 . Though L. rhamnosus belongs to the normal human rectal, oral, and vaginal mucosal flora, there are a few case reports of liver abscess due to L. rhamnosus, lactobacillemia, and infective endocarditis 40-46 . Lactobacillus sepsis has been documented in a few reports and was directly linked with the ingestion of probiotic supplements, especially among immunocompromised patients and those with endocarditis 40 . Kunz et al. 47  It has been suggested that the presence of a single major risk factor (immunocompromised state and premature infants) or more than one minor risk factor (cardiac valvular disease, central venous catheter, impaired intestinal epithelial barrier, administration of probiotics by jejunostomy, and probiotics with properties of high mucosal adhesion or known pathogenicity) merits caution in using probiotics because of the risk of probiotics-sepsis 58 .
Other safety concerns of theoretical importance are genetic transfer of antibiotic resistance from probiotic strains to more pathogenic bacteria in intestinal microbiota (particularly Enterococcus and Staphylococcus aureus) 59,60 , deleterious metabolic activities, and excessive immune stimulation in susceptible individuals 3,14 . Many strains of Lactobacilli are naturally resistant to vancomycin.

Future directions
As is evident from many recent studies, probiotics have a promising role in prophylaxis and the treatment of various conditions in critically ill children. However, these results are derived mainly from studies conducted in single centers and are limited by many factors, including small sample sizes, different populations and disease conditions studied, and heterogeneity in the probiotic strains, dose, and duration used. For probiotics to exert their action, it is important that they achieve tight adhesion to intestinal mucosa, and this may be difficult in critical illness. Most of the strains colonize the intestine only after 1 week of consumption, whereas early and effective mucosal adherence is needed to prevent MODS in critically ill children. Well-designed, large multi-center studies are needed for a better understanding of the role of probiotics in critically ill children as well as their pharmacokinetics, mechanisms of action, appropriate dose, administrative regimens, interactions, side effects, risk-benefit potential, and selection of specific probiotics (single-strain or multi-strain), dose, and duration for specific critical care conditions.

Conclusions
Probiotics have the ability to restore the imbalance of intestinal microbiota and function in critically ill children and have been used for various indications, including the prevention of AAD, HCAIs, VAP, Candida colonization, and invasive candidiasis. Safety may be of concern in critically ill, fragile children, as probiotic strains may (albeit rarely) cause bacteremia, fungemia, and sepsis. Welldesigned multi-center RCTs are needed to address these issues before the routine use of probiotics is recommended in critically ill children.
Author contributions Sunit C. Singhi conceived the plan of the review, drafted the broad outline, critically reviewed the draft, and finalized the manuscript. Suresh Kumar carried out the literature search and drafted the manuscript. Both authors read and approved the final manuscript.

Competing interests
The author(s) declared that they have no competing interests.

Grant information
The author(s) declared that no grants were involved in supporting this work.

Open Peer Review Current Referee Status:
Editorial Note on the Review Process are commissioned from members of the prestigious and are edited as a F1000 Faculty Reviews F1000 Faculty service to readers. In order to make these reviews as comprehensive and accessible as possible, the referees provide input before publication and only the final, revised version is published. The referees who approved the final version are listed with their names and affiliations but without their reports on earlier versions (any comments will already have been addressed in the published version).