![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Clinical Microbiology Reviews, October 2003, p. 658-672, Vol. 16, No. 4
0893-8512/03/$08.00+0 DOI: 10.1128/CMR.16.4.658-672.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Canadian Research and Development Centre for Probiotics, The Lawson Health Research Institute, London, Ontario N6A 4V2,1 Departments of Microbiology and Immunology,2 Surgery, University of Western Ontario, London, Ontario N6A 5C1, Canada3
SUMMARY INTRODUCTION PRESENT STATUS OF PROBIOTICS IN CLINICAL PRACTICE EVIDENCE OF PROBIOTIC EFFECTIVENESS Probiotics for Newborns and Children HOW PROBIOTICS REDUCE THE DURATION OF DIARRHEA ADDITIONAL APPLICATIONS OF ORAL PROBIOTICS Bacterial Gastroenteritis Helicobacter pylori Infections and Complications Inflammatory Diseases and Bowel Syndromes Cancer Mucosal Immunity Allergy WOMEN'S REPRODUCTIVE AND BLADDER HEALTH PROBIOTICS FOR SURGICAL INFECTIONS GUIDELINES FOR THE EVALUATION OF PROBIOTICS Strain Identification In Vitro and In Vivo Experiments Safety Phase 2 Clinical Studies Phase 3 Clinical Studies Health Claims and Labeling THE WAY FORWARD ACKNOWLEDGMENTS REFERENCES
| SUMMARY |
|---|
 Top NextReferences |
|---|
| INTRODUCTION |
|---|
|
TopPreviousNextReferences |
|---|
Probiotics represents an expanding research area. A Medline search of the term probiotics illustrates the significant increase in research undertaken in this area during the past 5 years: over 1,000 publications cited, compared to 85 for the previous 25 years. While this demonstrates the potential significance of this emerging field, much still remains to be done to standardize the meaning of the term probiotic and which strains actually fulfill the criteria of true probiotic microorganisms. In addition, although clinical evidence of the tangible benefits of probiotics is mounting, this does not yet reflect the commercial front. Unfortunately, many so-called probiotic products have not been properly identified, documented, manufactured under good manufacturing practices, or proven clinically, yet various companies make claims that lead consumers and caregivers to believe that they are using reliable products. Thus, the establishment of standards and guidelines represents a necessary first step in making sure that probiotic products are indeed legitimate and effective. Such standards and guidelines have recently been generated and will be presented later.
| PRESENT STATUS OF PROBIOTICS IN CLINICAL PRACTICE |
|---|
|
TopPreviousNextReferences |
|---|
Many health care professionals such as holistic practitioners, naturopaths, chiropractors, and herbalists routinely use products perceived to contain lactobacilli, bifidobacteria, and other possible probiotics. However, depending upon the training center, physicians may not be exposed to programs that discuss and evaluate the advantages and disadvantages of so-called nontraditional, complementary or alternative medicine, within which probiotics is sometimes placed. Governmental agencies such as the Food and Drug Administration are designed to separate and regulate drugs from other substances, which inadvertently makes it very difficult for small health companies to have the resources to seek claims and drug approval status for probiotics. Meanwhile, physicians rightly require that the medicines they prescribe or recommend have been tested, shown to have clinical effects, and be produced in reliable, reproducible product formulations. However, the current research-funding environment has not been conducive to sufficiently adequate testing of many probiotic strains in clinical practice.
Several factors are now leading many physicians to examine probiotics and other alternatives to pharmaceutical remedies. These include the surging levels of multidrug resistance among pathogenic organisms, particularly in hospitals, the increasing demands of consumers for natural substitutes for drugs, and the emergence of scientific and clinical evidence showing the efficacy and effectiveness of some probiotic strains. The FAO and WHO guidelines, albeit several years away from implementation in United Nations member countries, will ensure that reliable, clinically proven probiotics are available. Without such product formulations, physicians have little to offer their patients. Analyses of probiotic strains show that very few are currently available as drugs, foods, or dietary supplements.
| EVIDENCE OF PROBIOTIC EFFECTIVENESS |
|---|
|
TopPreviousNextReferences |
|---|
Bacterial colonization or infection of the intestine by pathogens such as Clostridium, Escherichia, Klebsiella, Salmonella, Shigella, Campylobacter, Pseudomonas, Streptococcus, Enterococcus, Staphylococcus aureus, and coagulase-negative staphylococci increases the risk of necrotizing enterocolitis. If nonpathogens, such as lactobacilli and bifidobacteria, colonize the intestine, or if breast milk rather than formula is used, the incidence of necrotizing enterocolitis has been reported to fall (73). At the time of this finding in 1990, the authors estimated that in British neonatal units, exclusive formula feeding could account for approximately 500 extra cases of necrotizing enterocolitis and 100 deaths each year. No recent comparison is available, but changes in infant formulas have been made over the past 13 years, so conclusions drawn from that time may or may not be relevant today.
Low-birth-weight premature infants delivered by caesarian section are often ill equipped for life outside the womb. They require intensive care, and for those who are breast fed, the feeding usually only begins several days after the infants are exposed to a plethora of microbes, many of which have pathogenic potential. This indicates that the normal process by which organisms such as lactobacilli are ingested via vaginal birth and propagated by the mother's milk do not take place. As a result, this may allow pathogens to establish within the premature intestine. Furthermore, infants given antibiotics at birth retain an abnormal microbiota 4 weeks later, such is the dramatic impact of these agents (44).
The intestinal microbiota in low-birth-weight premature infants can be dominated by many pathogens such as Enterococcus faecalis, Escherichia coli, Staphylococcus epidermidis, Enterobacter cloacae, Klebsiella pneumoniae, and Staphylococcus haemolyticus (44, 77). In particular, Clostridium perfringens has been isolated from 40% of babies with necrotizing enterocolitis, compared with 13% of controls (P = 0.03). However, in premature infants given breast milk, lactobacilli and bifidobacteria are present in a more diverse microbiota. For example, in a study of the enteric microbiota of 25 babies with necrotizing enterocolitis compared to 23 matched controls, lactobacilli were less common in the necrotizing enterocolitis babies (12% versus 48%, P = 0.006) (13). These findings suggested a correlation between the reduction of lactobacilli and the increased risk of necrotizing enterocolitis.
Other studies indicated that bifidobacteria not only colonized the gut of animals, possibly helping to exclude pathogens; they also reduced endotoxemia and appeared to modulate the inflammatory cascade (20). Perhaps the most impressive indication that probiotics could benefit newborns comes from a human trial with 2.5 x 108 live Lactobacillus acidophilus and 2.5 x 108 live Bifidobacterium infantis in 1,237 neonates in Colombia. Compared with 1,282 hospitalized patients seen during the previous year, treatment with these strains resulted in a 60% reduction in necrotizing enterocolitis and overall mortality (58). Although historical comparisons are not an ideal clinical trial design, the results nevertheless are striking and warrant consideration. A subsequent study involved newborn infants with a gestational age of <33 weeks or birth weight of <1,500 g and a standard milk feed supplemented with Lactobacillus sp. strain GG in a dose of 6 x 109 CFU once a day until discharge (
47 days) (30). The study found a reduced rate of necrotizing enterocolitis compared to placebo (1.4% versus 2.7%) but was not statistically significant, suggesting that either the GG strain is not as good as the L. acidophilus-B. infantis combination, milk is not an effective delivery system, or probiotics are not as effective as earlier thought (58). A further study of enteral feeding of premature infants with Lactobacillus sp. strain GG showed that the organism could be recovered from the stool and was thus delivered and survived passage, even though it did not appear to confer any detectable benefits (90).
Failure of the GG strain to prevent necrotizing enterocolitis does not necessarily indicate a lack of benefit to newborns. A double-blind, randomized, placebo-controlled trial involving 132 participants over a 2-year period showed that daily feeding of two capsules containing 1010 Lactobacillus sp. strain GG to pregnant mothers who had at least one first-degree relative (or partner) with atopic eczema, allergic rhinitis, or asthma and after birth to the mother and to the babies for 6 months significantly reduced the incidence of allergic atopic dermatitis (15 of 64 [23%] versus 31 of 68 [46%], P = 0.0008) (66). This implies a functional modulation of immunity rather than a specific antipathogen reaction in the gut. This effect has now been shown to remain at 4-year follow up (67).
The immune response within the gastrointestinal tract is a fine balance between the release of proinflammatory (e.g., interleukin-1, -6, and -8 and tumor necrosis factor) and anti-inflammatory (e.g., interleukin-1RA, -4, and -10) cytokines (75a, 135a). In a review on mucosal immunity starting at birth, Walker (146) reported a correlation between a normal gut microbiota and protection against various infections. This is an important observation because it supports the concept of early intestinal colonization with organisms such as lactobacilli and bifidobacteria and possibly subsequent protection from necrotizing enterocolitis and other diseases.
It is estimated that every 15 s a child dies from diarrheal disease somewhere in the world. In a study in 204 undernourished, 6- to 24-month-old children in Peru, once-daily intake of Lactobacillus rhamnosus GG 6 days a week for 15 months led to significantly fewer episodes of diarrhea (5.21 versus 6.02 episodes of diarrhea per child per year in the placebo group; P = 0.028) (96). However, this type of study is difficult to verify because there is little control over the organisms to which the children are exposed and the compliance in taking the treatment. At the least, probiotics provide a safe and potentially beneficial remedy, especially when delivered in milk, which provides the child with nutrition and a means to overcome adverse effects of fluid loss. Current WHO recommendations state that clinical management of acute diarrhea should include replacement of fluid and electrolytes losses along with nutritional support (150). As such, oral rehydration salts are widely used in diarrheal disease management.
The strongest evidence of a beneficial effect of probiotics has been established with L. rhamnosus GG and B. lactis BB-12 for prevention and L. reuteri SD2222 for treatment (51, 52, 62, 80, 122, 131, 138) of acute diarrhea mainly caused by rotaviruses in children (Table 1). The study designs cited are similar, randomized, double blinded, and placebo controlled. The statistically significant reduction in the duration of diarrhea is consistent and quite convincing, especially for the GG strain used in several of the trials. Note also that unlike many pharmaceutical treatments, there were no significant side effects reported. In a European study, faster hospital discharge was achieved in addition to improvement in clinical outcome (51). Two hundred ninety-one children 1 to 3 months of age were randomly allocated to receive oral rehydration solution plus placebo or 1010 L. rhamnosus GG. After rehydration in the first 4 to 6 h, patients were offered their usual feedings plus free access to the same solution until diarrhea stopped. Duration of diarrhea was reduced from an average of 3 days to 2.4 days (P = 0.03). In a randomized, placebo-controlled study of 40 patients between 6 and 36 months of age hospitalized with acute diarrhea (75% rotavirus), treatment with high doses of L. reuteri (assumed to be strain SD2222) (1010 to 1011 CFU) for up to 5 days resulted in reduction in duration of watery stools (1.6 versus 2.9 days in the placebo group) (P = 0.07) (131). In a second prospective, randomized, placebo-controlled trial of children between 6 and 36 months of age admitted for rotavirus-associated diarrhea, three groups received either 1010 or 107 CFU of L. reuteri SD2222 or a matching placebo once a day for up to 5 days. The outcome supported the earlier findings, with a mean duration of watery diarrhea being optimal for patients given the highest dose of lactobacilli (1.5 days versus 1.9 days for the lower dose versus 2.5 days for the placebo group) (132).
|
| HOW PROBIOTICS REDUCE THE DURATION OF DIARRHEA |
|---|
|
TopPreviousNextReferences |
|---|
The second potential mechanism may be that the immune response is enhanced by lactobacilli, leading to the observed clinical effect (65). This is supported by the protective effect which local immunoglobulin A (IgA) antibodies appear to confer against rotavirus (148). However, a problem with this theory is that given that diarrhea appears to cease within 1 to 3 days in patients who would otherwise suffer for 4 to 6 days, the lactobacilli would need to trigger the antibody response rapidly so that it interfered with further viral activity. Animal studies do indicate that secretory IgA can be triggered by lactobacillus ingestion (115), but the rate was not determined, nor was the influence on cessation of fluid loss across the secretory cell membranes. Modification of the cytokine profile to one that enhances anti-inflammatory cytokines (23) or attenuation of the virus's and/or toxin's effect on the enteric nervous system might provide rapid cessation of epithelial secretion and diarrhea. Alternatively, stimulation of T cells to produce gamma interferon, leading to potential inhibition of chloride secretion, might also inhibit diarrhea. One aspect of the immunity theory that needs to be clarified is why lactobacilli, which we assume are present in the child' intestine, appear unable to prevent infection, yet those administered orally thereafter help to clear the diarrhea.
A third mechanism could involve a signal(s) from lactobacilli to the host that downregulates the secretory and motility defenses designed to remove perceived noxious substances. Glycosylated intestinal mucins inhibit rotaviruses (151a), and MUC2 and MUC3 mRNA expression is increased in response to lactobacillus signaling, protecting cells against pathogenic bacterial adhesion (76). However, direct host cell signaling between lactobacilli and secretory cells has not yet been investigated. Attachment of the virus causes cytokine prostaglandin and nitric oxide release from the enterocytes, both of which could affect motility. The possibility exists that lactobacilli could alter this release (151).
A final theory is that lactobacilli produce substances that inactivate the viral particles. This has been shown in vitro (18), with supernatants from Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 inactivating 109 particles of the double-stranded DNA adenovirus and the negative-stranded RNA vesicular stomatitis virus within 10 min. The effect was likely due to acid, but more specific antiviral properties have not been ruled out. Whether or not viral killing activity can inhibit diarrhea remains to be determined.
More detailed mechanistic research is needed to understand how probiotic strains reduce the duration of diarrhea in conjunction with rehydration therapy. Such studies could lead to a better understanding of the dynamics within the intestinal microbiota that is being disrupted and depleted by rapid fecal loss. In doing so, new interventional therapies should be generated to quickly and effectively trigger the cessation of not only rotavirus illness but also other gastrointestinal infections that debilitate patients for 2 to 3 days.
| ADDITIONAL APPLICATIONS OF ORAL PROBIOTICS |
|---|
|
TopPreviousNextReferences |
|---|
A major problem associated with antibiotic use, particularly clindamycin, cephalosporins, and penicillins, is the onset of diarrhea, usually caused by Clostridium difficile. This organism is not uncommon in the healthy intestinal tract, but the disruption of the indigenous microbiota by antibiotics may lead to an abnormal elevation of C. difficile and subsequent symptoms related to toxin production (7). The acquisition of toxin-producing C. difficile is approximately 2,700 cases per 100,000 exposures to antibiotics in the community (8). In an effort to further refine standard antibiotic treatment of C. difficile infections, patients have been given standard antibiotic for 10 days plus either the yeast Saccharomyces boulardii (1 g/day for 28 days) or placebo. A significant decrease in recurrence was observed in patients treated with high-dose vancomycin (2 g/day) and S. boulardii compared with those who received high-dose vancomycin and placebo (16.7% versus 50%, P = 0.05) (137) (Table 2).
|
However, not all clinical trials have shown effectiveness. In one prospective, randomized, double blind, placebo-controlled trial, 302 hospitalized patients on antibiotics were randomized to receive Lactobacillus sp. strain GG (2 x 1010 CFU/day) or placebo for 14 days (143). Diarrhea developed in 39 (29.3%) of 133 patients in the GG group and 40 (29.9%) of 134 placebo patients (P = 0.93). Two potential flaws in this study are that only a few C. difficile infections were detected upon development of diarrhea and that patients self-reported stool consistency. Thus, the findings do not prove or disprove the ability of lactobacilli to treat antibiotic-associated diarrhea. In another study with the yeast probiotic S. boulardii, 69 patients over the age of 65 years prescribed antibiotics were randomized to receive either 113 g of S. boulardii twice daily or placebo for as long as they received antibiotics (71). There was no evidence that S. boulardii altered patients' bowel behavior or prevented the appearance of C. difficile toxin in the stool.
A meta-analysis to evaluate the efficacy of probiotics in prevention and treatment of diarrhea associated with the use of antibiotics recently showed an odds ratio of 0.39 (95% confidence interval, 0.25 to 0.62; P < 0.001) in favor of active treatment over placebo with S. boulardii and 0.34 (0.19 to 0.61; P < 0.01) for lactobacilli (33). The authors concluded that S. boulardii and lactobacilli have the potential to prevent antibiotic-associated diarrhea, but efficacy remains to be proven. Although such meta-analyses are useful, they become more credible if sufficiently large, similarly planned studies have been undertaken. In this meta-analysis study, there were only nine analyzable trials and four different probiotic strain combinations were used, emphasizing that more studies are required. It should be noted that cases of fungal infections have been reported following S. boulardii treatment, albeit in rare instances, usually associated with immunocompromised, catheterized patients (55).
In the second study (36), 53 patients infected with H. pylori were randomized to receive either 180 ml of L. johnsonii La1-acidified milk (LC-1) or a placebo twice a day for 3 weeks. All subjects also received clarithromycin (500 mg twice a day) during the last 2 weeks of acidified milk therapy. Esophagogastroduodenoscopy and biopsies were performed along with the urease test and histology. The LC-1 strain decreased H. pylori density in the antrum (P = 0.02) and the corpus (P = 0.04), as well as inflammation and gastritis activity in the antrum (P = 0.02 and P = 0.01, respectively). The unanswered question is whether the reduction is sufficient for short or long term clinical relief.
Not all efforts with probiotics have been successful, and one study that used yogurt to deliver a proposed probiotic organism failed to confer a beneficial effect (149). However, this study had several flaws; it assessed eradication only by a urea breath test, it used strains that had only inhibited H. pylori in vitro and were thus not proven probiotics, and it was only tested in 27 asymptomatic women. Further studies are therefore needed.
The study was very encouraging, and with this product already on the market, it will be interesting to see the outcome of clinical experience. In terms of product manufacturing, however, it is not clear how reproducible the preparations are in terms of the proportions of each strain or the effect each strain has in isolation as well as in combination. Mechanistic studies are needed to determine the product's effect on host cells, mucus, and immune defenses and the long-term impact on intestinal and fecal microbiota and patient outcome.
Based upon some animal experiments and pilot human studies (50), there is a potential therapeutic role for probiotics and prebiotics in patients with inflammatory bowel disease, including children. However, large, double-blinded controlled trials are needed to confirm efficacy and to document dosage and treatment parameters (91), and we must await the results of one such double-blind, placebo-controlled multicenter efficacy study under way to determine whether L. rhamnosus GG has any effect in children with Crohn's disease.
The demonstration that abnormal colonic fermentation occurs in some patients with irritable bowel syndrome has led to the suggestion that probiotic therapy might alleviate the condition. A double-blind, placebo-controlled, cross-over, 4-week trial of L. plantarum 299V in 12 previously untreated patients with irritable bowel syndrome showed no effect on the breath hydrogen test, gas production rate, or symptoms (126). This finding contradicts an earlier study with the same organism, in which 20 irritable bowel syndrome patients treated with this lactobacillus for 4 weeks reported resolution of their abdominal pain, compared to 11 patients on placebo (P = 0.0012) (93). An improvement in symptoms was noted in 95% of patients in the lactobacillus group versus 15% on placebo (P < 0.0001). Whether these positive results were in part due to patient selection or use of probiotics in liquid rather than dried form remains to be determined in a larger patient study. Suffice it to say, the role of probiotics in treating or preventing irritable bowel syndrome has not yet been clearly proven (78).
Dietary consumption of B. lactis HN019 and L. rhamnosus HN001 in randomized, placebo-controlled human studies showed measurable enhancement of immune parameters in the elderly (4, 130). The precise mechanisms of action remain to be established in patients, but the ability of Lactobacillus strains to activate macrophages (89) and stimulate secretory IgA and neutrophils without release of inflammatory cytokines (42, 65, 115, 133) could be important. It should also be recognized that not all immune effects are necessarily beneficial to the host. As Perdigon and Holgado (100) rightly point out, the major cell wall component of lactobacilli, muramyldipeptide, can be pyrogenic, and some Lactobacillus strains can enhance the Th1 proinflammatory pathway and would not be appropriate for mucosal immunity (75). The dosage and duration of therapy must also be considered so as to optimally enhance and not suppress immunity (99).
The use of lactic acid bacteria, such as L. plantarum NCIMB 8826, to deliver vaccines is being tested (121). This strain is being tested to deliver nontoxic tetanus toxin fragment C, and three types of constructs induced strong specific immune responses in mice. Future studies of this nature will explore the extent to which lactobacilli can immunize against various pathogens that attack the mucosal surfaces of the mouth, intestine, vagina, and respiratory tract.
As stated earlier, the application of probiotics to prevent allergic reactions became more prominent with the double-blind, randomized, placebo-controlled trial showing that L. rhamnosus GG given to pregnant women for 4 weeks prior to delivery and then to newborn children at high risk of allergy for 6 months caused a significant reduction in early atopic disease (66). Further clinical studies with L. rhamnosus GG and B. lactis BB-12 appear to have been useful in infants allergic to cow's milk (61, 79). The current understanding suggests that the probiotic organisms reverse increased intestinal permeability, enhance gut-specific IgA responses, promote gut barrier function through restoration of normal microbes, and enhance transforming growth factor beta and interleukin-10 production as well as cytokines that promote production of IgE antibodies (63, 66). The role of T-helper 1 (Th1) enhancement and T-helper 2 (Th2) reduction remains to be proven.
| WOMEN'S REPRODUCTIVE AND BLADDER HEALTH |
|---|
|
TopPreviousNextReferences |
|---|
Questions have been raised as to how abnormal is defined, given that subjects are not ill. The vaginal microbiota is often in a state of flux, as shown by Nugent score analysis, culture, and molecular tracking (17, 29, 32, 144). The Nugent score (94) is determined by microscopic analysis of vaginal cells collected from the vagina. When the field of view is dominated by lactobacillus morphotypes, the score is low (0 to 3), and when it is dominated solely by gram-negative rods (indicative of anaerobes like Gardnerella vaginalis or uropathogens like Escherichia coli) or gram-positive cocci like group B streptococci or enterococci, the score is high (8 to 10). Intermediate values indicate the presence of pathogens and lactobacilli in a sort of transition state. The factors that contribute to the transition from asymptomatic to symptomatic infection or a return to one that is healthy remain to be determined.
Nevertheless, the incidence of urinary tract infection, bacterial vaginosis, and yeast vaginitis, estimated to affect one billion women each year (the rate for urinary tract infection alone is 0.5 cases per person per year), means that the likelihood of infection is high. Indeed, the presence of pathogens dominating the vagina increases severalfold the likelihood that a woman will develop a symptomatic infection. In short, an abnormal microbiota may indeed lead to a symptomatic vaginal or bladder infection.
The concept of restoring the Lactobacillus content of the vaginal microbiota as a barrier to prevent infection was first conceived by Canadian urologist Andrew Bruce in the early 1970s. Extensive research since then has shown that certain Lactobacillus strains are able to colonize the vagina following vaginal suppository use (18) and reduce the risk of urinary tract infection, yeast vaginitis (108), and bacterial vaginosis (112). Strain selection at that time, and even recently, has been based upon in vitro tests and source of the strains (17, 111, 116, 144), but human studies provide the definitive answer to whether or not strains can function as probiotics.
A study on the prevention of urinary tract infection entailed once-weekly vaginal administration of a suppository containing 109 L. rhamnosus GR-1 and L. fermentum B-54 for 1 year and comparing the rate of urinary tract infection occurrence with that in the previous year in 25 women (108). There was a significant reduction in urinary tract infection during lactobacillus use (from an average of six episodes per year to 1.6 episodes per year [30%]; P < 0.0001). No side effects were reported. This compares favorably with two studies that used daily antibiotic therapy to prevent urinary tract infection. The first was a randomized study of 64 patients with a history of recurrent urinary tract infections given trimethoprim (100 mg) at night for 1 year or methenamine hippurate (1,000 mg) every 12 h or asked to cleanse the perineum (especially the periurethral area) twice daily with povidone-iodine solution (15). In this study, the urinary tract infection recurrence rates fell from 6 in the previous year to 2.1, 2.0, and 2.2, respectively, for the three groups.
The second study was an 18-year assessment of 219 women given one of three nitrofurantoin regimens daily for 1 year (14). The mean incidence of urinary tract infection fell from 6.9 per year to 1.3 per year. Notably, 14% of patients were allergic to the antibiotic, and 40% reported at least one adverse side effect, with nausea, gastrointestinal, genitourinary, and skin effects being the most common. In addition, 25.6% of 43 patients taking 50 mg of microcrystalline nitrofurantoin stopped prematurely as a result of an adverse event. Even the use of another antibiotic, cefaclor, 250 mg daily, in 37 patients resulted in an average of 2.4 breakthrough infections per year; 12.8% reported a side effect, and 7.7% stopped taking the drug (16). In short, Lactobacillus therapy taken once per week with no side effects resulted in as low a rate of urinary tract infection as several daily antibiotic regimens with numerous side effects. While a larger phase three trial has not been performed with Lactobacillus strains GR-1 and B-54, the phase two findings are worthy of note.
In terms of preventing or treating bacterial vaginosis, recent studies have shown that daily ingestion of capsules containing L. rhamnosus GR-1 and L. fermentum RC-14 by 19 women with a bacterial vaginosis microbiota resulted in a normal microbiota (by Nugent scoring) in 81% of cases, compared to 50% in women given placebo (P = 0.001) (109, 112, 114). This is not yet sufficient evidence to use oral probiotics for symptomatic bacterial vaginosis treatment, but it does illustrate the potential to reduce the incidence of recurrent bacterial vaginosis that is common after antibiotic treatment (54, 72). Another study, the results of which are only available on a web site, show that L. crispatus CTV05 given vaginally after metronidazole treatment for bacterial vaginosis resulted in a clinical cure at 30 days in subjects colonized by lactobacilli (70%) compared to noncolonized (47%) patients receiving placebo (P < 0.001). Further studies with vaginal Lactobacillus treatment of bacterial vaginosis that are more likely to deliver lactobacilli in higher numbers and more quickly than oral ingestion are warranted. Indeed, certain Lactobacillus strains, including L. crispatus CTV05, L. rhamnosus GR-1, and L. fermentum RC-14, are able to remain in the vagina for several months after insertion (18, 118).
The prevention or resolution of bacterial vaginosis is particularly important in women at risk of human immunodeficiency virus (HIV) infection. Studies have shown that women with bacterial vaginosis (no lactobacilli) are at significantly increased risk of HIV (127). Studies of 94 prostitutes in Madagascar showed bacterial vaginosis prevalence of 85% (9); a study in Malawi showed an odds ratio of 3.0 (95% confidence interval, 2.4 to 3.8) for an association between bacterial vaginosis and HIV (139); another Malawi study of 1,196 HIV-seronegative women showed that bacterial vaginosis was significantly associated with antenatal HIV seroconversion (adjusted odds ratio = 3.7) and postnatal HIV seroconversion (adjusted rate ratio = 2.3) (140); a cross-sectional study of 144 female commercial sex workers in Chiang Mai, Thailand, found a significant association between bacterial vaginosis and seropositivity for HIV (odds ratio, 2.7; 95% confidence interval, I0.3 to 5.0) (27); and a study of 4,718 women in Uganda showed an adjusted odds ratio of HIV-1 infection and bacterial vaginosis of 2.08 (95% confidence interval, 1.48 to 2.94) (127).
The depletion of lactobacilli and the risk of HIV was further illustrated in a study of 657 HIV-1-seronegative women in Kenya, where only 26% were colonized with Lactobacillus species, and the absence of these organisms was associated with an increased risk of acquiring HIV-1 infection (hazard ratio, 2.0; 95% confidence interval, 1.2 to 3.5) (82). The authors concluded that treatment of bacterial vaginosis and promotion of vaginal lactobacilli may reduce a woman's risk of acquiring HIV-1, gonorrhea, and trichomoniasis.
Having shown that certain Lactobacillus strains can colonize the vagina, this raises the questions of whether and how probiotics can reduce the risk of HIV infection (56). As stated earlier, supernatants from the strains L. rhamnosus GR-1 and L. fermentum RC-14 can inactivate viruses within minutes. It is presumed that a simple acidification of the vagina could affect HIV, but whether other mechanisms such as blocking receptor binding of the virus to CD4+ cells are at work remains to be investigated. Given the failure of current management and interventional steps to halt the AIDS epidemic, use of oral or vaginal lactobacilli appears worthy of consideration, especially given that they can be delivered relatively inexpensively to large numbers of people on the African continent, where drug supplies are often inaccessible or financially prohibitive.
Another potential application is for pregnant women to reduce the risk of bacterial vaginosis infection associated with infant mortality and preterm labor. Oral probiotics would be particularly useful in this case, as they can be administered safely during pregnancy (66). For reasons not yet known, some lactobacilli, such as L. rhamnosus GG and L. acidophilus NCFM, appear to be not well suited for the urogenital tract (18, 106), while products on the market such as the vaginal suppository Fermalac, comprising L. rhamnosus and other strains (Rosell, Montreal, Canada), have no peer-reviewed studies proving eradication of bacterial vaginosis. Thus, for clinical practice at present, there are few clinically proven, commercially available options to antibiotic and antifungal therapy for urogenital infections and antiviral drugs for HIV spread.
| PROBIOTICS FOR SURGICAL INFECTIONS |
|---|
|
TopPreviousNextReferences |
|---|
Application of probiotics for surgical patients is not necessarily limited to skin and wounds. In the first of three intriguing studies, L. plantarum 299 given with enteral fiber nutrition decreased the rate of postoperative infections in liver transplant patients at very high risk of infection, organ rejection, and death (105). These patients are immunosuppressed and malnourished, and the infecting organisms often originate in the patient's own intestine, spreading through translocation. Endotoxemia from gut pathogen overgrowth also leads to further complications. The lactobacilli in that study were administered 24 h after surgery four times a day for 6 weeks. Only 13% of patients developed infections, compared to 34 to 48% in controls, with a drop in enterococcal infections being notable. Interestingly, the drop in infections occurred for pneumonia, sepsis, and cholangitis as well as others. In a second paper by Rayes et al. (104), patients undergoing major abdominal surgery (such as resection of the liver, stomach, or pancreas) benefited from L. plantarum 299 in terms of fewer infections, fewer antibiotics prescribed, shorter hospital stay, and lower incidence of other complications. In the third randomized, double-blind study, patients with severe acute pancreatitis who received freeze-dried lactobacilli with oat fiber for 1 week had significantly fewer episodes (4.5% versus 30%) of infection and pancreatic abscesses (98).
Rather than use antibiotics in an attempt to "decontaminate" the intestine prior to surgery, these studies indicate that there is merit to administering probiotic organisms to reduce the risk of complications. Further studies are needed on other probiotic strains with different food additives and to determine whether some patients should not be given this therapy because of elevated risk of Lactobacillus-associated bacteremia (103).
| GUIDELINES FOR THE EVALUATION OF PROBIOTICS |
|---|
|
TopPreviousNextReferences |
|---|
|
The second consideration for particular strains that are being targeted for probiotic use is to have a clear and consistent strain designation, such as L. rhamnosus GG. This will allow physicians and consumers to track publications associated with that strain and to verify that the strain has indeed been shown to have probiotic benefits. An important example of why the failure to name strains leads to confusion is a publication which cited so-called probiotic yogurts as being responsible for side effects in patients (134). In fact, 12 of the 16 strains or products mentioned did not fit the definition of true probiotics (107). Likewise, claiming benefits on web sites or labels for strains that are either not in the product or have never been proven for that product formulation (60, 119, 152) is tantamount to the misrepresentation of facts. This is not to say that such products are necessarily inferior or unreliable; it's just that the producers need to prove their merit.
Of the other tests that require validation of in vivo performance, many still have merit in assessing and characterizing organisms as well as investigating potential mechanisms of action. For example, research that showed that lactobacilli could adhere to intestinal cells and signal mucus production which prevented pathogen adhesion (76) was extremely valuable and provided a new appreciation for function in the gut. Studies may also provide insight into how lactobacilli interfere with E. coli pathogenesis, including analysis of lipoteichoic acid antagonism of the inflammatory effect of lipopolysaccharide through competition with soluble CD14 (lipopolysaccharide binding protein) (145). This important concept shows how probiotic bacteria may directly produce benefits for the mammalian host.
The issue of safety becomes more complicated if one considers organisms such as Enterococcus spp. as probiotics (39). These bacteria are present in relatively high numbers in the intestine and are often included in so-called probiotic cocktails, particularly in animal feed. However, enterococci have emerged as an important cause of nosocomial infections, and isolates are increasingly vancomycin resistant (41, 53). The same is true of Saccharomyces boulardii, an organism used widely as a probiotic yet one which has been associated with episodes of fungemia (55).
A case has been made that Enterococcus and perhaps also S. boulardii not be referred to as probiotic for human use (38), but the onus should be on the producer that this and any other organism contemplated for human use not be a significant risk. For example, E. coli would not be regarded as a prime probiotic candidate because of so many of its strains are pathogenic, yet E. coli 83972 has been used quite effectively to prevent bladder infections in spinal cord-injured patients (31). That protocol was a prospective, nonrandomized, pilot clinical trial on 44 patients with spinal cord injury who had neurogenic bladder and had frequent episodes of symptomatic urinary tract infection. The bladders of the patients were inoculated with E. coli 83972, and among the 30 who became colonized with this organism there was a 63-fold reduction in the rate of symptomatic urinary tract infection versus the baseline prestudy period (mean, 0.06 versus 3.77 episodes of symptomatic urinary tract infections/patient-year, P < 0.001). In this example, safety did not appear to be an issue, and indeed the benefits outweighed the risks as far as could be determined. A more subtle form of safety may involve minimizing the transfer of drug resistance genes.
In order to establish safety guidelines for probiotic organisms, recognizing that many are Generally Recognized as Safe, the FAO and WHO recommended that probiotic strains be characterized at a minimum with a series of tests, including antibiotic resistance patterns, metabolic activities, toxin production, hemolytic activity, infectivity in immunocompromised animal models, side effects in humans, and adverse incidents in consumers (37). One possible scheme for testing toxin production has been recommended by the European Union Scientific Committee on Animal Nutrition (124). Given the rare incidence of side effects with Lactobacillus probiotics, large monitoring studies might prove useful. To date, there have been no reports of adverse overdose events caused by probiotics.
| THE WAY FORWARD |
|---|
|
TopReferences |
|---|
| ACKNOWLEDGMENTS |
|---|
G. Reid declares that he owns intellectual property rights associated with Lactobacillus strains GR-1 and RC-14.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
Top |
|---|
