Do probiotics have a therapeutic role in gastroenterology?

Abstract

Several hundred species of bacteria inhabit the gut, and affect its cell biology, morphology and homeostasis. Many bacteria are however potential pathogens, especially if the integrity of the epithelial barrier is physically or functionally breached. Conversely, the interaction between host and commensal microbes can confer important health benefits. This has led to commercial and public interest in 'probiotics', live microbes principally taken as food supplements. Might probiotics also be used in disease therapy Experimental evidence that probiotics modulate gut physiology, particularly barrier integrity and immunological function, underpins exciting new gastroenterological research. We discuss below the scientific basis for probiotic effects and present a critical perspective for their use in relation to gastrointestinal disease.

Key Words: Probiotics; Gastroenterology

INTRODUCTION

The concept of probiotics probably dates back to 1908, when Nobel Prize winner Eli Metchnikoff suggested that the long life of Bulgarian peasants resulted from their consumption of fermented milk products[1]. In 1965 Lilly and Stillwell first used the term ‘probiotic’ when describing ‘substances secreted by one organism which stimulate the growth of another’[2]. Parker[3] described probiotics as ‘organisms and substances which contribute to intestinal microbial balance’ and Fuller proposed in 1989 that probiotics were ‘a live microbial supplement which beneficially affects the host animal by improving its microbial balance[4]. Salminen et al[5] defined them as ‘foods containing live bacteria which are beneficial to health’, whilst Marteau et al[6] define them as ‘microbial preparations or components of microbial cells that have a beneficial effect to health and well being’. Such definitions underpin the current popular commercial usage of various ‘friendly bacteria’ to secure non-specific benefits to health.

With improved understanding of the physiology and therapeutic role of probiotics, definitions have evolved bolder claims, which now enter medical territory. Charteris et al[7] defined probiotics as ‘micro-organisms, which when ingested, may have a positive effect in the prevention and treatment of a specific pathological condition’.

Two related terms are prebiotics and synbiotics: prebiotics are defined as “non digestible food ingredients that beneficially affect the host by selectively stimulating the growth and/or activity of one or a limited number of bacteria in the colon that can improve host health”[8]. When prebiotics and probiotics are administered together, this is referred to as a synbiotic.

NORMAL GUT MICROFLORA

The colon contains tenfold more bacteria than the total number of mammalian cells constituting the host. The relationship is symbiotic: for example, bacterial vitamin K synthesis contributes to haemostasis, whilst short chain fatty acids generated by colonic bacteria salvage additional energy from otherwise ‘wasted’ dietary fibre.

Colonisation of the gastrointestinal tract starts immediately after birth, initially with maternal vaginal and intestinal flora. Other sources are diet (breast or formula based feeds)[9-12] and environment, as reflected by the different gut flora in infants born in developing and developed countries[13-15]. Infants who are breast fed predominantly harbour Bifidobacteria whilst formula fed infants have a more complex bacterial profile comprising Enterobacteria, Bacteroides, Clostridia, Lactobacilli, Bifidobacteria and Streptococci[11].

These ‘pioneer’ bacteria are important because they modulate epithelial cell gene expression, creating a favourable habitat for themselves by inhibiting the growth of bacteria introduced later[16]. This renders initial colonisation causally determinant to the final composition of bacterial flora in adults[17].

During development, the gut flora changes. The mouth harbours mainly facultative and strict anaerobes including Streptococci, Bacteroides and yeasts. The oesophagus has no significant resident microbial colonisation but is constantly transited by swallowed organisms. The widely held idea that the upper gut is largely sterile is not valid. The stomach and duodenum harbour up to 10⁴ colony forming units (CFU) per gm of Candida albicans, Bacteroides, Lactobacillus and Streptococcus. H pylori are specifically adapted for gastric residence. The jejunum again harbours Bacteroides, Candida albicans, Lactobacillus and Streptococcus but with a content of 10⁵-10⁷ CFU/g. Scepticism that probiotics will not survive the passage through the acidic stomach is therefore unfounded. From the ileum onwards bacterial colonisation increases from 10⁷-10⁸ CFU/g in the ileum to 10¹⁰-10¹¹ CFU/g in the colon, with a predominance of Bacteroides, Bacillus, Clostridium, Enterococcus, Peptostreptococcus, and Streptococcus species[18].

GUT MICROFLORA AND GUT PHYSIOLOGY

Recent research has demonstrated that a dynamic and reciprocal interplay exists between the gut microflora and the host. In particular, there is growing evidence that bacteria play an important role in directing epithelial differentiation and reinforcement of the gut barrier, through complex host-bacterial cross talk which occurs at a molecular level in interacting cells. This new biology underpins the putative effects of probiotics, since host-bacterial interactions can be re-engineered by purposeful manipulation of luminal ecology.

The physical barrier: bacteria and tight junctions

The intestinal epithelium constitutes an anatomical and functional barrier, effectively a bipolar monocellular obstacle between luminal microbes and the cells of the lamina propria. Barrier function is normally maintained by a complex interplay of numerous proteins, assembled to form the tight junction complexes (TJs)[19,20] in the juxta-apical region of the cell membrane (Figure 1). Commensal organisms contribute uncharacterised constitutive signals supporting epithelial integrity. Disruption of TJs may be elicited by pathogenic bacteria, stress and injury, via pro-inflammatory cytokines. Disruption of TJ integrity results in increased paracellular permeability, which is measurable in a variety of experimental models, and which may initiate, exacerbate or perpetuate intestinal inflammation in disease[21]. Altered intestinal permeability has been demonstrated in Crohn’s disease, celiac disease, intestinal infections and NSAID-induced enteropathy[22]. In order to counteract the harmful effects of luminal pathogens and toxins, and to protect barrier homeostasis, intestinal epithelial cells exhibit several additional defensive features, which include production of defence peptides and mucins[23]. In addition, a class of bacterial-sensing immunocytes, the dendritic cells, are able to project sensory dendrites into the lumen between adjacent enterocytes via TJ regions. Dendritic cells express receptors evolved to sense bacterial components, through which highly patterned host immune responses are evoked appropriately and constantly. Supporting experiments in rats have shown that colonisation of an excluded colonic loop with E. coli increased paracellular permeability, but this was partially reversed by colonisation with a putative probiotic, Lactobacillus brevis[24]. In addition, the chronic colitis occurring in Interleukin-10 (IL-10) deficient mice[25] is associated with increased colonic permeability, which is reversed in mice pre-treated for 4 wk. with a probiotic product, VSL#3 consisting of Bifidobacterium, Streptococcus and Lactobacillus species.

Figure 1 Tight junctions have a ‘barrier’ function in epithelia.

The potential cellular basis has also been addressed in human cell models[26]. The influence of whole probiotic bacteria E. coli Nissle 1917 (EcN) or VSL#3, bacterial cell lysates or conditioned medium from bacterial cultures were assessed against a variety of barrier and defensive parameters in human intestinal epithelial cell lines. These included a measure of TJ function (transepithelial resistance, TER) and tight junction protein abundance, IL-8 secretion and mucin gene expression. In addition, protective effects on pathogen (Salmonella dublin) induced alterations were analyzed. The probiotic mixture and soluble protein released from it, increased basal TER, prevented pathogen-induced decrease in TER, and stabilized TJs. This suggested that the organisms, or their secretory products, functionally modulate the intestinal epithelium of the host. These include competition of organisms for contact with the epithelial surface, stabilization of the cytoskeletal and barrier function, and the induction of mucin gene expression. Gram-negative and gram-positive organisms differed in the cellular mechanisms activated: perhaps a combination of organisms might be more effective than the application of a single strain[26]. The stumbling block inevitably lies in translating observations in these reductionist models to a proven clinical effect.

Recent interest has centred on regulation of the barrier via epithelial sensing of microbial components. The Toll-like receptors (TLRs) are a class of pattern-recognition receptors that specifically discriminate between self and microbial non-self based on the recognition of molecular patterns[27]. TLRs thereby play an important role in immune responses and the induction of antimicrobial effector pathways, leading to elimination and exclusion of host-threatening pathogens. It has been shown that the intestinal epithelium expresses several TLRs, which include TLR2, TLR3, TLR4 and TLR5[24-34]. It is possible that certain probiotic agents contain TLR-specific immunostimulatory features, leading to the amelioration of colitis by restoring intestinal epithelial barrier to protect the host[35].

The functional barrier: probiotics and mucosal immunity

The gastrointestinal mucosa is the primary interface between the external environment and the immune system. In the complete absence of intestinal microflora antigen transport is increased, indicating that the normal gut microflora maintain gut defences[25]. The microflora affect the development of gut associated lymphoid tissue at an early age by directing the regulation of systemic and local immune responsiveness, including promoting tolerance via hypo-responsiveness to antigens from micro-organisms and food[36]. Probiotic organisms have been shown to modulate immunoglobulin production. Secretory IgA plays an important role in mucosal immunity, contributing functionally to the barrier against pathogenic bacteria and viruses[37-40]. An enhanced IgA immune response has been shown in children with Crohn’s disease treated with Lactobacillus GG[41]. Interestingly, Madsen et al[42] demonstrated that IL10 deficient mice displayed significantly higher basal numbers of adherent bacteria compared with healthy control mice. When the colon was repopulated with Lactobacillus reuteri enemas the proportion of adherent and translocated bacteria, and the development of colitis, was significantly decreased. Further, Schultz[43] demonstrated that feeding Lactobacillus plantarum attenuated established colitis in IL-10 knockout mice. In rats, the effects of Lactobacilli and fibre (oatbase) were studied in Methotrexate-induced enterocolitis. Rats received an intragastric infusion of an elemental diet, with or without supplementation of oatbase, Lactobacillus reuteri R2LC and Lactobacillus plantarum DSM9843[44]. Methotrexate was injected intraperitoneally on d 3. By d 6 Lactobacillus decreased intestinal inflammation, re-established intestinal microecology and reduced bacterial translocation to extra-intestinal sites.

Data from human studies support a role for the gut microflora in the development of several gut associated inflammatory conditions, most likely triggered by immune response to their antigenic structures[45]. Thus probiotics may exert clinical effects by altering the intestinal inflammatory response to the luminal microflora.

Trophic and nutritional effects of gut microflora

Probiotic organisms exert a potentially positive effect on gut function through a trophic action on gut mucosa. In experimental models, crypt cell turnover is reduced in the colon of rats bred in germ free environments (gnotobiotic animals). Germ free crypts also contain fewer cells than those of colonised rats[46]. Butyrate is an important source of energy for colonocytes[47], whilst acetate and propionate are found in portal blood and are eventually metabolised in the liver or peripheral tissues, in particular muscle[47,48]. The most important role of SCFA is probably their trophic effect on colonic epithelium. Short chain fatty acids stimulate epithelial cell proliferation and differentiation in large and small bowel in vivo[49]. Butyrate however inhibits cell proliferation in epithelial cell lines of neoplastic origin[50]. Further, butyrate is pro-apoptotic and promotes reversion of cells from neoplastic to non-neoplastic phenotypes[51]. SCFA generation can clearly be altered by manipulating colonic micro-ecology, so probiotics and prebiotics may find a role in the prevention of colonic neoplasia and in the therapy of inflammatory bowel diseases.

Colonic micro-organisms play an important role in vitamin synthesis[52,53] and in the absorption of calcium, magnesium and iron[54-56]. Ion absorption in the caecum is improved by carbohydrate, and production of short chain fatty acids particularly acetate, propionate, and butyrate.

Interactions with mucus

A mucus gel covers the gut epithelium acting as a protective barrier against pathogens and reducing physical trauma. Change in the mucus content or structure compromise barrier function. Interactions occur between bacteria and mucus, including the probiotic bacteria that bind to intestinal mucus[57]. This is potentially advantageous since it inhibits adhesion of enteropathogenic bacteria to mucus. For example Enterococcus faecium inhibits the adhesion of enterotoxigenic E. coli K88 to porcine small intestine mucus[58].

PROBIOTICS IN CLINICAL GASTROENTER-OLOGY

The experimental data discussed above demonstrate that several ‘probiotic’ organisms exert biological effects which might translate into clinical benefits. Current clinical evidence is limited and non-uniform. Key observations in relevant conditions are discussed below.

Probiotics in GI infections

Probiotics have an emerging role in the treatment of gastrointestinal infections. Probably the best described is in acute infantile diarrhoea. Lactobacillus strain GG in fermented milk or freeze- dried powder was shown to reduce the duration of diarrhoea in acute rotavirus infection compared to a placebo group given pasteurised yoghurt[59]. Other studies have confirmed these results[60,61]. Suggested mechanisms of action are stabilisation of indigenous microflora[62], reduction in the increased gut permeability caused by rotavirus infection[63] and reduction in the duration of virus shedding[64]. This may be cause-specific. In a multicentre European trial of probiotics in acute childhood diarrhoea caused by rotavirus and other pathogens[65], probiotics (Lactobacillus GG) shortened the duration of diarrhoea in rotavirus diarrhoea, but showed no such effect with other pathogens.

Probiotics may also have a role in the prevention of acute infantile diarrhoea. In a double blind, placebo controlled trial hospitalised infants were randomised to receive a standard infant formula alone or supplemented with Bifidobacterium bifidum and Streptococcus thermophilus. After a 17-mo follow up period, 7% of those receiving a probiotic had experienced diarrhoea compared to 31% given the standard formula[66]. Viral shedding was lower in the probiotic supplemented group. Prophylactic use of Lactobacillus GG in 204 undernourished children followed up for 15 mo also decreased the incidence of acute diarrhoea[67]. This effect was confined to non-breast fed infants. In a more recently reported double blind randomised placebo controlled trial, 81 children between 1-3 years of age, hospitalised for reasons other than diarrhoea, and were given Lactobacillus GG or placebo for the duration of their admission. The incidence of nosocomial diarrhoea was lower in the probiotic group (7% as compared to placebo, 33%)[68]. Furthermore, although the prevalence of rotavirus infection was similar in both groups, the risk of rotavirus gastroenteritis was lower in the probiotic group. Finally, the benefits of Lactobacillus GG in rotavirus associated diarrhoea mainly comprise a reduction in duration of diarrhoea by 1-2 d compared to a median of 3 d.

The positive results for Lactobacillus GG in acute viral diarrhoea cannot necessarily be extrapolated to other probiotic strains, nor to other causes of acute diarrhoea. The data imply species and disease specificity, For example, in a study comparing various probiotic strains, Lactobacillus GG was found to have a beneficial effect in rotavirus gastroenteritis but this was not shared by L. rhamnosus, L. delbrueckii or Streptococcus thermophilus[60]. Saccharomyces boulardii has also displayed beneficial effects in acute diarrhoea in children and adults[60,70]. Enterococcus SF68 in adults with acute diarrhoea has however shown inconsistent results[71-73]. It is unclear whether the modest benefits suggested by these studies justify routine use of probiotics in diarrheal illnesses since most acute diarrhoeal diseases are self limited. The benefit may be greatest in situations where patients are at risk for complications such as in children with malnutrition in developing countries.

Antibiotic-associated diarrhoea

Diarrhoea can occur as an adverse effect of antibiotic therapy, in up to 39% of antibiotic treated hospitalised patients[74]. Broad spectrum antibiotics are more commonly implicated, possibly because of more profound alteration in colonic flora[75]. Several placebo-controlled studies have shown a decrease in the incidence of diarrhoea or change in stool consistency when patients were treated with probiotics in addition to antibiotics[76-82]. The probiotic organisms studied were Lactobacillus spp., Enterococcus and Saccharomyces boulardii. Not all studies support this possibility. A recent study of 267 patients on antibiotics randomised to Lactobacillus GG or placebo failed to show any decrease in incidence of diarrhoea, with 29% in both groups developing symptoms[83]. Three other smaller studies were also negative[78,84,85]. Although most studies looking at probiotics in antibiotic-associated diarrhoea are placebo controlled and conducted on a reasonable number of subjects, different antibiotics were used in these studies contributing to their heterogeneity. Two recent meta-analyses (of nine and seven randomised placebo controlled double blind trials) have looked at the effect of probiotics in prevention of antibiotic-associated diarrhoea[86,87]. The meta- analysis by D’Souza et al[87]reviewed nine randomised, double blind, placebo controlled trials. Four trials used yeast (Saccharomyces boulardii); four used Lactobacilli and another used a strain of enterococcus that produced lactic acid. Three trials used a combination of probiotic bacteria. Antibiotics were given with probiotics (or with placebo, in the control group) in all nine trials. The combined odds ratio in favour of active treatment over placebo in preventing diarrhoea associated with antibiotics was 0.37 (95% confidence interval, 0.26 to 0.53; P < 0.001). In the meta- analysis by Cremonini et al[86], twenty two studies using Lactobacillus and Saccharomyces species, matched the inclusion criteria of which seven studies were homogenous. The combined relative risk was 0.399 (95% confidence interval, 0.27-0.57) suggesting a strong benefit of probiotic administration on antibiotic associated diarrhoea. The pooled results suggested that probiotic administration had an overall benefit. However, the published data is discordant in that it is unclear what the optimal dose and timing of supplementation should be.

Clostridium difficile associated diarrhoea (CDAD)

Clostridium difficile is a Gram positive bacterium that can cause colitis, mediated by two enterotoxins, enterotoxin A and B. The pathophysiology is not fully clear but risk factors include intercurrent or continued antibiotic therapy, elderly age, renal disease and female sex[88,89]. Probiotics have been partially evaluated in the prevention of CDAD. Early uncontrolled trials using Lactobacillus GG[90-92] and a preliminary report of a controlled trial using Lactobacillus GG suggested benefit in recurrent CDAD[93]. Similarly uncontrolled or open label studies[94,95] and subsequently two controlled trials[96,97] have suggested efficacy of Saccharomyces boulardii in recurrent CDAD. In the study by McFarland et al[94], 124 patients were studied, including 64 patients with an initial episode of CDAD and 60 who had at least one previous episode of CDAD. Subjects received oral Saccharomyces boulardii (1 g/d for 4 wk or placebo) and were followed up for an additional 4 wk after therapy. Multivariate analysis showed that patients treated with S. boulardii and standard antibiotics had a significantly lower risk of CDAD (relative risk 0.43; 95% confidence interval, 0.20-0.97) compared with placebo. In their subsequent study, Surawicz et al[97] tested patients receiving a standard antibiotic for 10 d and then added S. boulardii (1 g/d for 28 d) or placebo. A significant decrease in recurrence of CDAD (16.7%) was observed in patients treated with high-dose Vancomycin (2 g/d) compared with those receiving Vancomycin and placebo (50%; P < 0.05). Most studies were small and were mostly not placebo controlled.

A recent systematic review looking at studies in which probiotic therapy was used for prevention and treatment of C. difficile- associated diarrhoea concluded that the evidence for routine clinical use of probiotics in this setting was insufficient[98]. Again, although probiotics are generally considered safe, case reports have described Saccharomyces cerevisiae fungaemia and deaths in immunocompromised and critically ill patients who received a commercial preparation of S. boulardii (genomically identical to S. cerevisiae)[99]. Their routine use can therefore not be recommended.

Probiotics in Inflammatory Bowel Diseases

Although the etiology of inflammatory bowel diseases is not entirely clear, dysfunctions in both innate and acquired immunity are implicated. There has been an increased interest in pathogenic and endogenous intestinal flora, with supportive data derived from several animal models. As noted above spontaneous colitis may develop in mice deficient in the immunoregulatory cytokine IL-10 but IL-10 deficient germ free mice remain disease free[100-102].

Clinical studies suggest a significant role for bacteria in the pathogenesis of human IBD. Crohn’s disease activity has been shown to improve with antimicrobial therapy, faecal diversion[103,104], bowel rest and intestinal lavage[105]. Furthermore, antibiotics may reduce postoperative relapse[106], postoperative pouchitis[107] and fistula related complications[108]. There has been particular interest recently in polymorphisms in the CARD15/NOD2 gene, an intracellular bacterial pattern recognition receptor, as a risk factor for the development of Crohn’s disease[109]. Probiotics may therefore exert benefits in IBD management by modulatory effects on intestinal flora. For example Lactobacillus GG, when administered to children with Crohn’s disease, increased mucosal IgA levels[41] improved intestinal permeability and reduced disease activity[110]. Further, the relapse rate in 32 adults with inactive Crohn’s disease was reduced to 6% when subjects in remission were treated with mesalazine and S. boulardii as compared to 38% with mesalazine alone[111]. However, in a placebo controlled study of 37 Crohn’s disease patients treated with Lactobacillus GG for 12 mo after curative resection, the probiotic did not prevent relapse: in fact more severe endoscopic findings were reported in the Lactobacillus group[112].

Recent clinical studies have evaluated the effect of non-pathogenic E. coli strain Nissle 1917 versus mesalazine for maintenance of remission in ulcerative colitis. Kruis et al[113] studied 120 and Rembacken et al[114] 116 patients with inactive ulcerative colitis over a period of 12 wk and 1 year respectively given either E. coli strain Nissle 1917 (MutaflorR) or mesalazine. These unblinded studies found a similar relapse) in both groups (73% in the mesalazine group and 67% in the E. coli group), suggesting that probiotic therapy may be an alternative maintenance therapy. Similar results were suggested in a third RCT study[115]. 327 patients were treated with mesalazine or E. coli Nissle 1917 for twelve months. Relapse rates were similar (45.1% in the probiotic group versus 36.4% in the mesalazine group). In a controlled trial by Tursi et al[116], low dose balsalazide with VSL#3 was shown to be more effective than balsalazide or mesalazine alone in patients with acute mild to moderate ulcerative colitis. The combination of a prebiotic and a probiotic (Bifidobacterium longum/Synergy 1) was associated with improvement in histological scores and measures of immune activation in a randomised controlled pilot study[117].

A recent study investigated the expression and function of CARD15/NOD2 in intestinal epithelial cell lines CARD15/NOD2 mRNA was expressed in both intestinal epithelial cell lines and primary intestinal epithelial cells. CARD15/NOD2 mRNA and protein were up-regulated by tumor necrosis factor alpha (TNF alpha) in SW480 cells. This study suggests that CARD15/NOD2 may serve as a key component of innate mucosal responses to luminal bacteria as an antibacterial factor[109]. Failure in this activity may contribute to the development of Crohn’s disease.

Probiotics in pouchitis

Pouchitis is an inflammation of an ileal reservoir surgically constructed in the management of IBD. It is associated with reduced counts of Bifidobacteria and Lactobacilli with increased numbers of Clostridia and anaerobes in faecal samples[118]. Increases in bile acids and decreases in short chain fatty acids, with a net increase in pH, may also be seen[118]. In a double blind randomized placebo controlled trial a probiotic mixture VSL#3 was studied over 9 mo in 40 patients with chronic relapsing pouchitis[119]. Relapse was defined by a pouch disease activity index (PDAI)[120] of 2 points or more and confirmed by endoscopy and histology, and was, strikingly, only identified in 15% of the VSL#3 group as against 100% in the placebo group. In a prophylactic study, 2 of 20 patients (10%) receiving 1 packet of VSL#3 1 year developed pouchitis, versus 40% of placebo treated patients[121].

Somewhat different conclusions were reached in two recently published studies[122,123]. In a group of 36 patients with recurrent or refractory pouchitis who had required antibiotics at least twice in the past year, patients were randomly assigned to VSL#3 or placebo after achieving remission, once daily for a year. More patients remained in remission in the probiotic group (85% vs 6%)[122].

In an observational study involving 31 patients treated with VSL#3 after achieving remission with Ciprofloxacin, only a minority of patients remained on probiotic therapy and in sustained remission, having stopped them due to recurrence of disease or adverse effects[123]. In summary then, the benefit of probiotics in Crohn’s disease remains unproven. The benefit of probiotics in ulcerative colitis remains unproven. E. coli Nissle 1917 appears promising and may be of value in patients intolerant of or resistant to 5-ASA preparations. Limited data from small controlled studies would suggest that VSL#3 is a reasonable therapy in the primary and secondary prevention of pouchitis.

Probiotics in critical illness

Some advocates propose that probiotics have an important emerging role in managing critical illnesses originating in the gastrointestinal tract. In a recent study from Hungary patients with severe acute pancreatitis were randomised upon arrival to hospital to receive one week of treatment with a twice daily administration of a freeze dried preparation containing 10⁹ live L. plantarum 299 with a substrate of 10 g oat fibre, or a similar preparation containing Lactobacillus which had been inactivated[124]. The study was stopped when the infection rate showed a significant difference in the two groups. This occurred when 45 patients had completed the study, 22 had received treatment with live and 23 with heat killed L. plantarum 299. Infected pancreatic necrosis occurred in 1 out of 22 subjects (4.5%) in the treatment group as against 7 of 23 (30%) in the heat killed group. The length of hospital stay was shorter in the live LAB group but it did not reach statistical significance due to small sample size[124].

Studies on cirrhotic patients have shown a decrease in the incidence of encephalopathy. A Chinese study recently reported 55 patients with minimal hepatic encephalopathy who were randomised to receive a synbiotic preparation, fermentable fibre alone or placebo. Synbiotic treatment was associated with a reduction in serum ammonia, endotoxaemia, reversal of encephalopathy and improvement in Child-Turcotte-Pugh score in 50% patients[125]. Following liver transplantation, bacterial and fungal infections may occur in the first month despite extensive antibiotic treatment and selective digestive tract decontamination. A German study showed a reduction in post-transplant infective complications using probiotics[126]. Another recent study looked at the effects of a synbiotic preparation on gut barrier function and in critically ill patients admitted to the Intensive Care Unit (ICU). Ninety patients admitted to ICU were randomised (45 in each group) to receive either synbiotic preparations (Lactobacillus acidophilus La5, Bifidobacterium lactis Bb12, Streptococcus thermophilus and Lactobacillus bulgaricus with oligofructose (as a prebiotic) or placebo. Patients in the symbiotic group had a lower incidence of potentially pathogenic bacteria (43% vs 75%, P = 0.01) and multiple organisms (39% vs 75%, P = 0.01) but there were no significant differences in both groups in terms of intestinal permeability, septic complications or mortality[127]. It should be pointed out that the study lasted only one week.

Probiotics and colon cancer

Colon cancer is one of the leading causes of death in the Western world. Dietary intake of red meat is probably associated with a higher risk whereas the intake of fruit, vegetables, fish and calcium are arguably associated with a lower risk. It is interesting that colon cancer risk is also lower in countries such as Netherlands and Finland where a larger quantity of yoghurt is consumed[128,129]. Could diet exert its effects via the microflora If so, mechanisms involved might include altered metabolic activity of the intestinal microflora, binding and degradation of potential carcinogens, production of antitumorigenic or mutagenic compounds and enhancement in host immune response. In animals where colon cancer was induced by chemical carcinogens, administration of lactic acid bacteria resulted in a suppression of DNA damage, tumour formation and growth[130-133].

One bio-epidemiological study showed a higher risk of colon cancer in the simple presence of Bacteroides species, but lower risk with Lactobacillus acidophilus, Lactobacillus S06 and Eubacterium aerofaciens identifiable in faecal flora[134]. This raises the intriguing hypothesis that similarly shifting the resident bacterial populations may be accompanied by parallel reductions in neoplastic risk. However, biodiversity in the colon may merely represent an epiphenomenological consequence of the dietary and environmental risk factors listed above.

Probiotics and digestion

Probiotics may have a promising role in certain aspects of human digestion as illustrated by some interesting studies. Lactose digestion has been shown to improve in lactose malabsorbers who consume live yoghurt rather than milk[135,136]. The beneficial effects of yoghurt in lactose malabsorbers may result from improved digestion of lactose in the colon from stimulation of colonic bacterial activity by lactic acid bacteria[137].

Probiotics in irritable bowel syndrome

Irritable bowel syndrome (IBS) is a collection of functional gastrointestinal symptoms such as abdominal pain, defaecatory frequency and or constipation. This area is inevitably contentious, since it remains unclear how much intrinsic intestinal pathology exists in IBS. However, changes in gut sensitivity and defaecatory function are clearly present. Alterations in the composition of intestinal flora have been reported but not proven including a decrease in faecal Lactobacilli, E. coli and Bifidobacteria[41,110,111] and an increase in other faecal anaerobes[138-140]. Symptomatic improvement was noted in a very small crossover trial of 18 patients with IBS given L.acidophilus[141] and in an uncontrolled study using E. faecium PR88[142]. On the other hand, no improvement was seen in bloating, pain or urgency to defaecate after the consumption of Lactobacillus GG for 8 wk[143,144].In a study reported by Saggioro et al[145], 50 patients with IBS according to RomeIIcriteria were randomly assigned to a probiotic preparation (a combination of Lactobacillus plantarum LPO 1 and Bifidobacterium breve Bro or placebo for 4 wk). Pain and severity scores decreased significantly after 14 d of treatment. In a more recent study, 77 patients were randomly assigned to a malted drink containing Lactobacillus salivarus UCC4331 or Bifidobacterium infantis 35 624 or a malted drink alone[146]. Significant improvement in symptoms was noted in the B. infantis group. A corresponding normalisation in the ratio of IL-10/IL 12 was also noted suggesting that the probiotic may help reduce a proinflammatory state associated with IBS.

However the heterogeneity of the various studies makes it difficult to draw conclusions on the effect of probiotics in IBS, and the field is bedevilled by the fact that all therapeutic interventions in IBS produce a 30%-50% placebo response.

Probiotics and Helicobacter pylori

H pylori infection is associated with gastritis, gastro duodenal ulcers and gastric malignancies. The majority of H Pylori hosts become hypochlorhydric with time. Clinical studies and experimental models have shown that the secreted products of Lactobacillus acidophilus can suppress H Pylori growth in vitro and in vivo L. johnsonii[147] and LG21[148] are effective in suppressing the growth of H pylori and reducing gastric inflammation. Placebo controlled studies have demonstrated a reduction in side effects of standard triple therapy if probiotics were administered concurrently[149-151]. Daily intake of inactivated L. acidophilus was shown to improve the efficacy of eradication treatment[152]. Only one study[153] showed that supplementation with fermented milk, containing live special probiotic L. casei DN-114001, confers an enhanced therapeutic benefit on H pylori eradication in children with gastritis on triple therapy. The theory that probiotic therapy enhances the disappearance of H pylori does not gain any strength from the available literature. Further clinical studies would be needed to evaluate the effects of long term ingestion of probiotics in preventing Helicobacter-associated diseases, but are unlikely to supplant H pylori eradication which is rapidly and highly effective.

In conclusion probiotics are live microbial food supplements or components of bacteria which may have beneficial effects on gastrointestinal health. Innate bacterial floras clearly play an important role in reinforcement of the physical gut barrier, affecting paracellular permeability, mucosal trophic action and microbiological interactions with mucosal lining of the gut. The key and unanswered question is whether the deliberate manipulation of the bacterial complement in the gut can confer clinical benefit. Probiotics do now appear to have a potential role in the prevention and treatment of various gastrointestinal illnesses, but it is likely that benefits achieved are specific to the bacterial species used and to the underlying disease context. Much further work is required from bench to bedside before we can realise the potential of these new interventions.