点击显示 收起
1 From the Gastroenterology Department, Hôpital Europèen Georges Lompidou, Assistance Publique des Hôpitaux de Paris and Paris V University, Paris, and the Institute of Physiology and Biochemistry of Nutrition, Federal Dairy Research Center, Kiel, Germany.
2 Presented at the symposium Probiotics and Prebiotics, held in Kiel, Germany, June 1112, 1998. 3 Address correspondence to P Marteau, Service de Gastro-entérologie, Hôpital Europèen Georges Lompidou, Lorue Leblanc, 75908 Paris Cedex 15, France.
ABSTRACT
Probiotics are nonpathogenic microorganisms that, when ingested, exert a positive influence on the health or physiology of the host. They can influence intestinal physiology either directly or indirectly through modulation of the endogenous ecosystem or immune system. The results that have been shown with a sufficient level of proof to enable probiotics to be used as treatments for gastrointestinal disturbances are 1) the good tolerance of yogurt compared with milk in subjects with primary or secondary lactose maldigestion, 2) the use of Saccharomyces boulardii and Enterococcus faecium SF 68 to prevent or shorten the duration of antibiotic-associated diarrhea, 3) the use of S. boulardii to prevent further recurrence of Clostridium difficileassociated diarrhea, and 4) the use of fermented milks containing Lactobacillus rhamnosus GG to shorten the duration of diarrhea in infants with rotavirus enteritis (and probably also in gastroenteritis of other causes). Effects that are otherwise suggested for diverse probiotics include alleviation of diarrhea of miscellaneous causes; prophylaxis of gastrointestinal infections, which includes traveler's diarrhea; and immunomodulation. Trials of gastrointestinal diseases that involve the ecosystem are currently being performed, eg, Helicobacter pylori infections, inflammatory bowel disease, and colon cancer.
Key Words: Probiotics bifidobacteria lactobacilli intestinal infections antibiotic-associated diarrhea gastroenteritis traveler's diarrhea intestinal flora inflammatory bowel disease colon cancer
INTRODUCTION
Probiotics can be defined as nonpathogenic microorganisms that, when ingested, exert a positive influence on the health or physiology of the host (1). They consist of either yeast or bacteria, especially lactic acid bacteria. Their fate in the gastrointestinal tract and their effects differ among strains (2). The effects of probiotics can be direct or indirect through modulation of the endogenous flora or of the immune system (2). Many health claims have been made concerning probiotics, especially concerning their potential to prevent or help cure intestinal disturbances; however, only a few probiotic strains were shown to be efficacious in randomized placebo-controlled clinical trials. In this article, we summarize the present knowledge on the therapeutic effects of probiotics in human gastrointestinal diseases.
IMPROVED LACTOSE DIGESTION AND OTHER DIRECT ENZYMATIC EFFECTS
Lactose maldigestion occurs frequently, especially in adults (primary lactose maldigestion) and in persons with bowel resection or enteritis (secondary lactose maldigestion). It is well established that persons with lactose maldigestion experience better digestion and tolerance of the lactose contained in yogurt than of that contained in milk (3). The mechanisms involved have been extensively investigated. The importance of the viability of lactic acid bacteria was speculated as pasteurization reduced the observed digestibility. At least 2 mechanisms, which do not exclude each other, have been shown: digestion of lactose in the gut lumen by the lactase contained in the yogurt bacteria (the yogurt bacteria deliver lactase when lyzed by bile acids) and slower intestinal delivery or transit time of yogurt compared with milk (36). In clinical practice, the replacement of milk with yogurt or fermented dairy products allows for better digestion and decreases diarrhea and other symptoms of intolerance in subjects with lactose intolerance, in children with diarrhea, and in subjects with short-bowel syndrome (3, 4, 7, 8). An enhanced digestion of a sucrose load was shown in infants with sucrase deficiency when they consumed Saccharomyces cerevisiae, ie, a yeast that contains the enzyme sucrase (9). This is yet another example of a direct effect of a probiotic; however, its relevance in the treatment of sucrase deficient subjects is not established.
ANTIBIOTIC-ASSOCIATED DIARRHEA
Diarrhea occurs in 20% of patients who receive antibiotics. Antibiotic-associated diarrhea (AAD) results from a microbial imbalance that leads to a decrease in the endogenous flora that is usually responsible for colonization resistance and to a decrease in the fermentation capacity of the colon. Clostridium difficile and Klebsiella oxytoca contribute to the occurrence of AAD in some cases and play a role in the pathogenesis of colonic lesions. Several attempts have been made to determine whether the administration of probiotics would prevent antibiotic-associated intestinal symptoms (mainly AAD). Randomized controlled trials that showed a significant therapeutic effect of probiotics are shown in Table 1; the effects of probiotics on C. difficile and K. oxytoca are shown in the next section. Three randomized, double-blind, placebo-controlled studies showed that oral administration of Saccharomyces boulardii (Ultralevure, Biocodex, France) can decrease the risk of AAD (Table 1). Another study showed that S. boulardii significantly shortened the duration of AAD (22). The mechanism involved is unclear because multiple biological effects of the yeast in the gastrointestinal tract have been shown, which may contribute to the clinical efficacy of S. boulardii (ie, effects against the population levels of C. difficile, toxins, and intestinal secretion) (23, 24). The therapeutic efficacy of other probiotics is not as well established. It is possible that differences in probiotic preparation may explain why a mixture of freeze-dried lactobacilli significantly prevented diarrhea in 1 study but not in 2 other studies (10, 11, 25; Table 1). Whether yogurt may help to prevent or cure AAD was suggested in open trials but has not been studied in controlled experiments (2).
View this table:
TABLE 1. Randomized controlled trials showing a significant therapeutic effect of probiotics in the prevention of antibiotic-associated intestinal symptoms (mainly diarrhea)
GASTROENTERITIS
Gastroenteritis is the main cause of acute diarrhea and is a frequent disorder that usually heals spontaneously within a few days. Gastroenteritis can be due to several viral or bacterial pathogens or to parasites, but the most frequent cause in children is rotavirus infection. The use of oral rehydration solutions is the main treatment, but it does not shorten the duration of diarrhea.
Curative treatment
Several controlled randomized trials showed a beneficial effect of probiotics and fermented dairy products in infantile or, less often, adult gastroenteritis; however, this is not a general property of all probiotics (2628). Lactobacillus rhamnosus GG (L. GG, Valio, Finland) has been shown to be effective in the treatment of infant rotavirus diarrhea (Table 2). L. rhamnosus GG repeatedly reduced the duration of diarrhea by about half in randomized controlled trials (Table 2). It also proved effective in the treatment of acute diarrhea in children in Asia (34, 35). Guandalini et al (38) recently reported the results of a double-blind multicenter European trial in children with acute diarrhea. Two-hundred eighty-seven children aged 136 mo with acute diarrhea were enrolled; they received oral rehydration solution formulated according to usual recommendations in addition to L. rhamnosus GG [109 colony-forming units (CFU)/250 mL] or placebo. The duration of diarrhea was 58 ± 28 h in the L. rhamnosus GG group and 72 ± 36 h in the placebo group (NS); diarrhea was significantly reduced by L. rhamnosus GG in children with rotavirus infection (56 ± 17 h compared with 77 ± 42 h; P < 0.05) but not in the 186 children who were rotavirus negative (59 ± 33 h compared with 69 ± 22 h). Administration of L. rhamnosus GG also shortened the duration of the hospital stay and the course of weight gain (38). The results of one study suggested that heat-inactivated L. rhamnosus GG was as effective as living L. rhamnosus GG in reducing the duration of diarrhea; however, the effect of the living probiotic was more pronounced on rotavirus specific immunoglobulin A response (45). Enterococcus faecium SF 68 (Bioflorin, Giuliani, Switzerland) was shown to significantly shorten the duration of diarrhea in 4 randomized controlled trials, 2 in infants and 2 in adults (Table 2). Other probiotics are probably also effective (Table 2). Boudraa et al performed a randomized study of yogurt compared with a milk formula in 112 young Algerian children with acute diarrhea (data not published). Both formulas were comparable in terms of lactose content, pH, flavor, and texture. The mean duration of diarrhea was significantly reduced from 65 ± 5 h in the milk group to 44 ± 5 h in the yogurt group. At 48 h, 35% of children in the milk group were cured of their diarrhea, compared with 64% in the yogurt group. The difference was even more pronounced when only the 72 infants with rotavirus were considered: 27% were cured with milk, compared with 68% with yogurt (G Boudraa, unpublished observations, 1996). Note that a significant shortening of gastroenteritis was reported in adults treated with heat-killed lactobacilli (Lacteol fort, Lactéol du Dr Boucard, France) (46).
View this table:
TABLE 2. Randomized controlled trials showing a significant therapeutic effect of probiotics to shorten the duration of acute gastroenteritis
Prevention
Several nonrandomized trials suggest a preventive effect of some fermented products on the risk of diarrhea in children (2, 47). Saavedra et al (44) showed that feeding Bifidobacterium bifidum and Streptococcus thermophilus to infants admitted to the hospital significantly reduced the risk of diarrhea and the shedding of rotavirus (Table 2). In a double-blind placebo-controlled trial, 55 children admitted to a chronic medical care unit were randomly assigned to receive a standard formula or a standard formula with B. bifidum and S. thermophilus. During follow-up, diarrhea occurred in 7% of the children receiving the probiotic and in 31% of the control subjects (P = 0.035), and shedding rotavirus occurred in 10% of children compared with 39% (P = 0.025), respectively.
INTESTINAL INFECTIONS AND COLONIZATION BY PATHOGENIC BACTERIA
The protective effects of probiotics against intestinal infections were shown in animal models (23, 24, 47). Mechanisms that may be implicated include the production of acids, hydrogen peroxide, or antimicrobial substances; competition for nutrients or adhesion receptors; antitoxin actions; and stimulation of the immune system.
Open trials suggested that some probiotics may help to eradicate pathogens in chronic carriers of salmonella and campylobacter (48, 49). Several reports related to patients experiencing a recurrence of C. difficile infections. This serious clinical problem occured in 20% of the subjects treated for a first episode of infection with this microorganism and in >40% of subjects who experienced several episodes. Several open studies performed in a limited number of subjects suggest a beneficial role of L. rhamnosus GG, S. boulardii, and Lactobacillus plantarum LP299v during C. difficilerelated infections (5056). Although these studies suggested a therapeutic effect, especially because they pertained to subjects with recurrent infection, they did not have the proof level of randomized controlled trials.
McFarland et al (57) performed a study that included 124 patients who were randomly asigned to receive a standard antibiotic treatment combined with either S. boulardii (1 g/d for 28 d) or a placebo. The risk of clinical recurrence for the subjects who had experienced several episodes of C. difficile infection was significantly reduced in the S. boulardii group: 34.6% compared with 64.7% in the placebo group (P = 0.04). The administration of L. rhamnosus GG to preterm infants hospitalized in a neonatal intensive care unit was attempted to decrease the risk of K. oxytoca colonization but was ineffective (58).
Helicobacter pylori would be a good target for an efficient probiotic therapy. Colonization of the gastric mucosa is strongly associated with gastritis, duodenal and gastric ulcers, and some malignancies. Antagonistic actions of some Lactobacillus strains against H. pylori in vitro were reported (59). Attempts to eradicate H. pylori in vivo with a probiotic have failed until now (60). However, a significant reduction of urease activity was reported in patients treated with a supernatant of Lactobacillus johnsonii LA1 (Nestlé, Switzerland Lausanne) associated with omeprazole (61).
TRAVELER'S DIARRHEA
Acute diarrhea occurs in about half of travelers who visit high-risk areas. Although most cases are mild and self-limiting, there is a considerable morbidity. Antibiotics are effective prophylaxis but are not recommended for widespread use (62, 63) and there is thus a need for cost-effective alternative treatments. Several studies were performed with the use of probiotics (Table 3). Some studies that used lactobacilli had negative results, whereas 4 studies that used diverse probiotics reported positive results (Table 3). Black et al (67) treated 94 Danish tourists participating in a 2-wk trip to Egypt with a mixture of Lactobacillus acidophilus, Lactobacillus bulgaricus, bifidobacteria, and S. thermophilus or a placebo in a randomized study. The frequency of traveler's diarrhea was reduced from 71% (very high) to 43% (P < 0.001). In a double-blind randomized study, Oksanen et al (69) reported a reduction of diarrhea by L. rhamnosus GG administration to subjects traveling to Turkey; however, the effect was significant for only one destination in Turkey. Another study used the same strain in 400 American travelers who were randomly assigned to receive L. rhamnosus GG or a placebo (70). One hundred fifty-five travelers were excluded, mainly because they did not take the medication. When only the subjects who took the capsules were considered, the risk of having diarrhea on any given day was 3.9% for patients treated with the probiotic compared with 7.4% in those not treated (P = 0.05). Kollaritsch et al (68) used S. boulardii in a double-blind placebo-controlled trial in which only 1016 of 3000 Austrian travelers were compliant. The protection against the occurrence of diarrhea was mild but significant and was dose-dependent (68; Table 3).
View this table:
TABLE 3. Randomized controlled trials of probiotics to prevent traveler's diarrhea
IRRITABLE BOWEL SYNDROME AND VARIOUS CONDITIONS WITH DIARRHEA
Some probiotics, including acidophilus or bifidus milk, were reported to relieve constipation in a short series of patients (2); however, these studies were not controlled. In a randomized placebo-controlled study including only 34 patients, Maupas et al (71) observed that S. boulardii decreased functional diarrhea but did not influence other symptoms of irritable bowel syndrome. Halpern et al (72) suggested in a randomized, double-blind, crossover trial that administration of heat-killed lactobacilli for 6 wk was more efficient than was placebo in relieving symptoms of irritable bowel syndrome. However, only 18 of 29 randomly assigned subjects were studied and this poor compliance was a weakness of that study. Hentschel et al (73) assessed the efficacy of 2 probiotic preparations containing lactobacilli and Escherichia coli (Hylac and Hylac N forte, Merckle, Blaubeuren, Germany) in 126 subjects suffering from nonulcer dyspepsia and did not observe any amelioration.
S. boulardii decreased the duration of diarrhea induced by tube feeding in 3 trials (7476). The most recent study was double-blind and compared the administration of 2 g S. boulardii/d with placebo in 128 critically ill tube-fed patients (76). Treatment with the probiotic reduced the percentage of days patients experienced diarrhea from 18.9% to 14.2% (P = 0.007). Two open studies proposed that lactobacilli might have some efficacy against small intestinal bacterial overgrowth (77, 78), but S. boulardii was ineffective in the only randomized placebo-controlled study (79).
Diarrhea is a nearly constant adverse effect of irradiation of the pelvis. A randomized controlled study by Salminen et al (80) showed a significant decrease in diarrhea in patients receiving L. acidophilus NDCO 1748 during pelvic irradiation. Previous open trials suggested the efficacy of freeze-dried lactic acid bacteria cultures for the same indications (80). Such potentially interesting therapeutic effects should be studied more thoroughly. Elmer et al (81) reported that high doses of S. boulardii might be effective in some subjects with HIV-related chronic diarrhea; however, further evaluation is warranted before firm conclusions can be drawn.
WELL-BEING
Compared with the numerous studies in patients, there have been only a few investigations of otherwise healthy people with or without mild gastrointestinal symptoms, and the often-claimed improvement of well-being by probiotics has not been proven until now.
In a recent controlled, randomized, double-blind study (de Vrese and Schrezenmeir, unpublished observations, 1998), 66 healthy, lactose-tolerant adults in 3 groupsafter a 3-wk preperiod without fermented foodconsumed 125 g/d of a chemically acidified milk product without bacteria (control) or with 2 strains of probiotic Lactobacillus (1010 CFU/d). Gastrointestinal symptoms and well-being were recorded by validated questionnaires and expressed as a sum score of 5 characteristics concerning intestinal function and pain. Within 1 wk, both probiotics, but not the artificially acidified milk product without bacteria, improved well-being and decreased gastrointestinal symptoms, from 6 to 4 points (on a scale of 030 points). These differences were significant (P < 0.05) with respect to both the control subjects and the preperiod without probiotics. This was the first time that such an effect was observed in healthy persons.
INFLAMMATORY BOWEL DISEASE
Inflammatory bowel disease refers to disorders of unknown cause that are characterized by chronic or recurrent intestinal inflammation. Such disorders include ulcerative colitis, Crohn disease, and pouchitis. The mechanisms responsible for initiation and perpetuation of the inflammatory process remains unknown, but the main theory is that inflammatory bowel disease may result from abnormal host responses to some members of the intestinal flora or from a defective mucosal barrier (82, 83). Treatment may be difficult and there is a need for new treatments to decrease the occurrence of symptoms and to prevent recurrence.
Several studies showed interesting effects of probiotics on inflammatory bowel disease in animals. Intracolonic administration of L. reuteri R2LC to rats with acetic acid-induced colitis significantly decreased the disease, whereas Lactobacillus HLC was ineffective (84). Administration of Lactobacillus reuteri R2LC and Lactobacillus plantarum DSM 9843 to rats with methotrexate-induced enterocolitis was associated with low intestinal permeability, bacterial translocation, and plasma endotoxin concentrations compared with rats with enterocolitis and no treatment (85). A few studies were also performed in patients. In an open study, a 10-d administration of L. rhamnosus GG to 14 children with active or inactive Crohn disease resulted in an increase in immunoglobulin Asecreting cells to ß-lactoglobulin and casein, which indicates an interaction between the probiotic and the local immune system (86). The lactobacilli did not influence the disease activity, however, because the study group was too small and the study was too short to assess accurately a clinical effect (86). Plein and Hotz (87) performed a pilot, double-blind, controlled study of the efficacy of S. boulardii on symptoms of Crohn disease. Twenty patients with active, moderate Crohn disease were randomly assigned to receive either S. boulardii or a placebo for 7 wk in addition to the standard treatment. A significant reduction in the frequency of bowel movements and in disease activity was observed in the group receiving S. boulardii but not in the placebo group.
Two studies (88, 89) compared the efficacy of an oral E. coli preparation [E. coli strain Nissle (Mutaflor, Ardeypharm GmbH, Herdecke, Germany)] and mesalazine (ie, the standard treatment) in maintaining remission of ulcerative colitis. The first study included a total of 120 patients with inactive ulcerative colitis. After 12 wk, 11.3% of the subjects treated with mesalazine had relapsed, compared with 16% of those treated with the probiotic. The second study included 116 patients and also showed that the probiotic preparation was as effective as mesalazine in inducing remission and preventing relapse (89). Several studies are currently testing the effects of probiotics on inflammatory bowel disease in Europe (90).
COLON CANCER
The endogenous flora and the immune system play a role in the modulation of carcinogenesis. Both may be influenced by probiotics and this has led to trials investigating the role of probiotics in preventing or curing tumors in animals (91). Several authors showed that some probiotics may decrease the fecal concentrations of enzymes, mutagens, and secondary bile salts that may be involved in colon carcinogenesis (91). Some but not all epidemiologic studies also suggest that consumption of fermented dairy products may have some protective effect against large colon adenomas or cancer (92). It is thus impossible to draw any conclusion at this time. Clinical studies are currently ongoing in Europe to study the effects of probiotics and prebiotics in subjects with colonic adenomas.
CONCLUSIONS
The concept of probiotics may occasionally favor overestimation of effects; however, accumulating research evidence suggests that probiotics may have a role in human therapies. In our opinion, the proven medical indications of probiotics for gastrointestinal disturbances are the following: 1) replace milk with yogurt in subjects with lactose intolerance, 2) use freeze-dried S. boulardii or E. faecium SF 68 to prevent AAD, 3) use freeze-dried S. boulardii to prevent further recurrence of relapsing diarrhea because of C. difficile, and 4) use fermented milk containing L. rhamnosus GG to shorten the duration of the diarrhea during rotavirus enteritis in children. Many other potential applications exist, but more controlled studies are required.
REFERENCES