Prebiotics have attracted interest for their ability to positively affect the colonic microbiota composition, thus increasing resistance to infection and diarrhoeal disease. This study assessed the effectiveness of a prebiotic galacto-oligosaccharide mixture (B-GOS) on the severity and/or incidence of travellers' diarrhoea (TD) in healthy subjects.
The study was a placebo-controlled, randomized, double blind of parallel design in 159 healthy volunteers, who travelled for minimum of 2 weeks to a country of low or high risk for TD. The investigational product was the B-GOS and the placebo was maltodextrin. Volunteers were randomized into groups with an equal probability of receiving either the prebiotic or placebo. The protocol comprised of a 1 week pre-holiday period recording bowel habit, while receiving intervention and the holiday period. Bowel habit included the number of bowel movements and average consistency of the stools as well as occurrence of abdominal discomfort, flatulence, bloating or vomiting. A clinical report was completed in the case of diarrhoeal incidence. A post-study questionnaire was also completed by all subjects on their return.
Results showed significant differences between the B-GOS and the placebo group in the incidence (P<0.05) and duration (P<0.05) of TD. Similar findings occurred on abdominal pain (P<0.05) and the overall quality of life assessment (P<0.05).
Consumption of the tested galacto-oligosaccharide mixture showed significant potential in preventing the incidence and symptoms of TD.
Travellers' diarrhoea (TD) is one of the most common conditions experienced by individuals travelling to a developing country (Kamat and Mathur, 2006). It has an attack rate of 20–50%, and it has been estimated that the disorder can affect over 11 million people annually (DuPont and Capsuto, 1996; Peltola and Gorbach, 2001). In 90% of patients who contract TD, symptoms occur within the first 2 weeks of travel (Steffen et al., 1983). A recognized definition of the syndrome of TD is accepted as ‘three to four unformed stools in 24 h and at least one of the following symptoms such as abdominal pain, nausea, vomiting, fever, blood or mucus in stools’ (Peltola and Gorbach, 2001). Studies have shown that bacteria and some viruses are the most common causative agents of TD (Black, 1990; Paredes et al., 2000) with entero-toxigenic Escherichia coli being reported as the bacterial cause in 50% of TD cases (Vila et al., 2000).
Management of TD has been so far based on the treatment of the condition using antibiotics and anti-diarrhoea medications such as fluoroquinolones, ciprofloxacin, rifaximin and loperamide (DuPont, 2001). However, these agents can potentially also have an impact on the commensal gut microbiota composition. Thus, alternative or supplementary approaches of preventative nature would be of great interest.
Bifidobacteria, together with lactobacilli, has an important function in the eco-physiology of the colonic microbiota. These organisms have been linked to increased resistance to infection and diarrhoeal disease (Tojo et al., 1987; Saavedra et al., 1994). The concept of using prebiotics for the management of gastrointestinal microbiota and function has attracted recent interest. Prebiotics are ‘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, and which may thus improve the health of the host’ (Gibson and Roberfroid, 1995). Prebiotics have been also reported to induce anti-microbial effects principally through their selective stimulation of indigenous beneficial genera (principally bifidobacteria), which also may be anti-pathogenic (Kunz et al., 2000; Gibson et al., 2005).
Among the various oligosaccharides tested for potential prebiotic application, galacto-oligosaccharides (GOS) have been seen to be efficacious. A recent study showed that a particular prebiotic GOS (B-GOS) was effective at increasing bifidobacterial numbers (Tzortzis et al., 2005) in the colonic content of piglets and at inhibiting the attachment of entero-hepatic E. coli (P<0.01) and Salmonella enterica serotype Typhimurium (P<0.01) to HT29 cells in vitro. The prevention of S. enterica serotype Typhimurium colonization and invasion by the specific prebiotic ingredient was also confirmed in an in vivo oral challenge study in mice (Searle et al., 2009). The prebiotic and bifidogenic properties of B-GOS were further confirmed in a randomized, double-blind, crossover, placebo-controlled intervention study in 59 healthy human adult volunteers (Depeint et al., 2008), as well as in a double-blind, crossover, placebo-controlled human study in 44 healthy elderly subjects (Vulevic et al., 2008). In the latter study, B-GOS was also reported to positively affect the immune function of the volunteers by increasing phagocytosis and NK cell activity.
These properties of the specific prebiotic mixture (increase in beneficial bacteria, stimulation of immune function and prevention of attachment and invasion of pathogenic bacteria) offer potential for the use of B-GOS in the prevention and/or reduction of bacterial-caused TD.
Materials and methods
The study was placebo-controlled, randomized, double blind of parallel design and the target enrolment was healthy volunteers, who would travel and stay in a country of low or high risk for TD for a minimum of 14 days and a maximum of 60 days. High-risk destinations for incidence of TD included Asia, The Middle East, Africa, Mexico, Central and South America, whereas low-risk areas included Turkey, South Africa and The Caribbean Islands, as determined by the Center for Disease Control and Prevention (CDC, 2008). Recruitment of subjects was conducted through advertisement of the study in the local and national press. Respondents to the advertisement were contacted by telephone and the study was explained to them. A series of screening questions were asked about their holiday, past and current medical conditions, travel details and TD history and concomitant medication, if any. Inclusion and exclusion criteria are given in Table 1. Prohibited medication during the study period included anti-depressants, narcotic analgesics, anti-cholinergic, anti-spasmodic agents for bowel diseases, immunosuppresive drugs and anti-tumour necrosis factors. However, antibiotics, anti-diarrhoeals (such as loperamide) and laxatives were allowed in the case of significant infections as ‘rescue’ medication. Subjects who met the inclusion criteria were sent a volunteer information leaflet and asked to read it thoroughly before taking part. Those who agreed to participate were then randomized into groups with an equal probability of receiving either of the two treatments. This ensured that numbers of subjects allocated to each treatment group were equal after every block of subjects. After allocation of the participant's number, they were sent a volunteer pack, which included volunteer information leaflet (for reference), consent form, diary card, clinical report form (to be filled in case of diarrhoeal incidence), instructions of how to consume the product, an explanatory letter for immigration purposes and the test product or placebo. The product used in this study was B-GOS (Bimuno, Clasado Ltd, Milton Keynes, UK), served in sachets (5.5 g), once daily. B-GOS is a GOS mixture already marketed in Europe and is generally recognized as safe, as defined by United States of America Food and Drug Administration (FDA, 2008). The GOS content of B-GOS is 48% (w/w) and its composition, in terms of degree of polymerization and saccharide linkages have been reported earlier (Vulevic et al., 2008). The placebo was maltodextrin (oligosaccharides without prebiotic effect) served in sachets (5.5 g), once daily. The interventions were in white aluminium foil sachets blind code labelled. A total of 30 such sachets contained in a blank paper box showing all the necessary information such as description of use and researcher contact details were provided to the volunteers at the beginning of the study.
Seven days before reaching their final destination, volunteers started consuming the product on a daily basis and kept a daily diary noting the number of bowel movements and average consistency of the stools (formed, loose, water or bloody), as well as occurrence of abdominal discomfort, flatulence, bloating or vomiting. The volunteers continued consuming the treatment while on their holidays. The outcome measures that were used for the holiday period, related to the following:
Number of bowel movements
Stool consistency (1=formed, 2=loose, 3=watery, 4=bloody)
Abdominal pain (0=none, 1=present but well tolerated, 2=present and interfering with but not preventing daily activities, 3=preventing normal daily activities)
Bloating (0=none, 1=present but well tolerated, 2=present and interfering with but not preventing daily activities, 3=preventing normal daily activities)
Flatulence (0=none, 1=present but well tolerated, 2=present and interfering with but not preventing daily activities, 3=preventing normal daily activities)
Vomiting (0=absence, 1=presence)
In the case of diarrhoeal incidence, defined as three or more bowel motions (semi-solid, watery or bloody) per day, the volunteer completed a clinical report form describing the signs, symptoms and duration of incidence, hospital admission history, history of exposure and onset, and progress of the diarrhoeal incidence. Incidence of TD with respect to subjects' age, gender and place of travel in subjects taking B-GOS was compared with the placebo.
After completion of the study, volunteers were also asked to complete the short World Health Organization Quality of Life-BREF questionnaire (post-study questionnaire). This consisted of 22 questions concerning how they felt about their quality of life, health or other areas of their life during the last 4 weeks. A scoring system was introduced and scores for each group were calculated based on specific equations as described in World Health Organization Quality of Life-BREF (WHO, 2004). Questions concerned different aspects of subjects' quality of life during their vacation were divided into four domains. Domain 1 related to physical health (daily living activities, dependence on medical substances and medical aids, pain and discomfort, sleep and rest), Domain 2 mainly concerned their psychological condition (positive/negative feelings, self-esteem), whereas domain 3 looked at social relationships (personal relationships, social support). Last, domain 4 related to their overall feeling of health as influenced by the environment while being abroad (physical safety, participation in and opportunities for recreation/leisure activities).
Completed diary cards were collected for assessment along with any unused sachets. Adverse events were also addressed along with any concomitant medications used.
The procedures followed were in accordance with the ethical standards of the Research Ethics Committee of the University of Reading on human experimentation and with the Helsinki Declaration of 1975 as revised in 1983. Ethical approval for the study was granted by the Research Ethics Committee of the University of Reading in July 2007 (approval reference number: 07/21). The study lasted for 9 months.
For the power calculation, we predicted that 50% of the subjects would experience diarrhoea and that GOS may reduce this rate by 20%. For 80% power and 5% significance, this gave 93 subjects per treatment group (MGH Biostatistics Software, Massachusetts General Hospital, Boston, MA, USA). Statistical analysis was conducted using GENSTAT statistical package version 10.0 (VSN International Ltd, Hemel Hemstead, UK). A value of P<0.05 was taken to indicate statistical significance. Data were analysed by Fisher's exact text, t-test, ANOVA and multiple regression analysis. In the regression analysis, the dependent variable was one of the outcome measures mentioned above, and independent variables were included, corresponding to the outcome measure for the pre-holiday period. Thus, each subject's bowel score on holiday was analysed and compared between treatments, taking into account the corresponding pre-holiday scores. Other independent variables were included where appropriate, such as duration of travel, TD risk of holiday destination, recent travel history and recent TD.
A total of 201 subjects were recruited and randomized, but 42 subsequently did not complete the study, of whom 23 had been allocated to the placebo group and 19 the test (B-GOS) group. Figure 1 shows the total number of people who enrolled and finished the study and reasons for not completing. Five subjects did not complete because of adverse reactions and their main complaint was occurrence of mild abdominal pain, which was reported during the pre-holiday period.
The distribution of destinations was similar between placebo and prebiotic group (Table 2). In total, 159 subjects completed the study, of which 81 consumed the B-GOS powder and 78 were in the placebo group. Relevant information is summarized in Table 2 concerning the age group (average), sex and earlier history of TD.
Bowel habit data for the two time periods (pre-holiday, holiday) in the two groups based on the analysis of the received and completed diaries are given in Table 3 and statistically significant differences indicated. Regression analysis indicated that the number of bowel movements (average per day) significantly increased during the holiday compared with the pre-holiday period in the placebo group (P<0.05), but not in the B-GOS group (P=0.12). For stool consistency and abdominal pain, significant differences occurred between pre- and holiday periods in the placebo group (P<0.05), but not in B-GOS (P=0.10). This was not unexpected as diarrhoeal incidence and duration was significantly higher (P<0.05) in the placebo group, as later analysed below. However, there was no difference (P=0.23) between the two treatments during the holiday period. A similar analysis for flatulence and bloating showed that neither the study period nor treatment exerted a significant effect.
The analysis of clinical report forms concerning incidence of diarrhoea and related symptoms showed a significant effect of B-GOS on the incidence and duration of TD.
Data in Table 4 show the incidence and duration of diarrhoea as well as the duration of other symptoms such as abdominal pain, vomiting, fever, anorexia, headache and dizziness. Fisher's exact test indicated that a significantly (P=0.03) lower number of volunteers who consumed B-GOS (n=19) had diarrhoeal during the holiday period compared with the placebo group (n=30). Regression analysis similarly indicated that the duration of diarrhoea and associated abdominal pain was significantly lower in the B-GOS group when compared with placebo, regardless of destination or holiday duration. For the subjects experiencing diarrhoea, the quality of life questionnaire showed higher scores in the B-GOS group compared with the placebo group (P<0.05).
There were no other differences in other symptoms between the two treatment groups.
TD is a very common problem for those travelling overseas for either business or holidays. As the cause of acute diarrhoea includes infecting organisms, antibiotic prophylaxis is recommended as the main ‘rescue’ treatment. Despite this, 20–50% of travellers still experience diarrhoea, and new preventive measures are required.
The use of probiotic bacteria has been proposed as an alternative way to infection and diarrhoea disease prevention based on studies that have reported amelioration of attacks of diarrhoea in both children and adults (Cremonini et al., 2002; Huang et al., 2002). Depending on the specific strains used, these bacteria have been shown to either adhere to the small intestine and thus prevent the adhesion of pathogenic bacteria or colonize the large intestine and thus increase resistance to invading pathogens. However, the benefits of using probiotics in the prevention of TD have so far proven to be limited, probably because of strain stability and survivability. Prebiotics, as an alternative, are more stable than probiotics during passage through the upper gastrointestinal tract and are able to induce anti-microbial effects principally through their selective stimulation of indigenous beneficial gut genera (Vila et al., 2000; DuPont, 2001; Buddington et al., 2002). On the other hand, the fact that prebiotics act mainly in the large intestine, while the infective organism causing TD act in the small intestine, presents a significant limitation for their application as preventative treatment against acute diarrhoea, as has earlier been shown in a study using fructo-oligosaccharides (Cummings et al., 2001). To overcome this issue, in this study, we have decided to use an oligosaccharide mixture, B-GOS, which beyond its prebiotic properties has been reported to also directly interact with the host epithelium preventing the adhesion and invasion of gastrointestinal pathogens acting in the small intestine (Searle et al., 2009) as well as positively affecting the immune function of immune-compromised subject (Vulevic et al., 2008).
In this study, 31% of persons who took part reported diarrhoea during the holiday period. Of those, 62% were in the placebo group and 38% in the B-GOS group showing a statistically significant difference (P<0.05) between the two treatment. Moreover, the duration of diarrhoea for the people in the B-GOS group was 2.4 days compared with the 4.6 days in the placebo group (P<0.05), even though 34% of volunteers in the placebo group followed a ‘relief’ treatment (such as loperamide or ciprofloxamin) versus 22% in the B-GOS group. The overall average bowel movement number of the participants was not significantly different during the holiday period between treatments, but it should be noted that the baseline (pre-holiday) number was significantly higher for the B-GOS group when compared with placebo, indicating substantial heterogeneity among the interventions as far as the average daily number of bowel movements was concerned.
In an earlier study, the use of another prebiotic (fructo-oligosaccharides) showed a significant effect in flatulence, which significantly increased in the fructo-oligosaccharides group (Cummings et al., 2001). Although this is expected for prebiotic oligosaccharides because of their physiological and metabolic properties, in this study the use of B-GOS did not show any significant effect on flatulence. The lack of flatulence as a side effect for B-GOS has earlier been reported in a clinical study with irritable bowel syndrome sufferers (Silk et al., 2009), during which administration of B-GOS, at similar levels as used in this study, significantly reduced the reported flatulence and bloating scores of the patients. This property of B-GOS has been attributed to the selective fermentation of the oligosaccharide mixture by Bifidobacterium spp., which do not produce gas, as a result of its mode of preparation.
Analysis regarding the overall quality of life of the volunteers, as based on a scoring system, showed that although there were no differences between the treatments for the subjects that did not experience diarrhoea, the people enrolled in the B-GOS group reported better (P<0.05) overall feelings concerning their health than those in the placebo group during the duration of acute diarrhoea.
In summary, the results of this study is that a significantly lower number of subjects who consumed the B-GOS treatment experienced diarrhoeal episodes during their holiday period compared with the placebo group. Similarly, the duration of diarrhoea and abdominal pain was significantly lower in the B-GOS group compared with the placebo (P<0.05) resulting in the volunteers reporting better quality of life during this experience. Although further studies are required to understand the exact mechanism of action, this beneficial effect of B-GOS, which has not been reported for other prebiotic oligosaccharides, is attributed to the increased functionality of the specific mixture, which includes, besides the prebiotic properties in the large intestine, direct protection from the adhesion and invasion of pathogenic organism in the small intestine and stimulation of the immune function resulting in increased levels of phagocytic and NK cell activities.
Conflict of interest
Dr Tzortzis is employed by Clasado Ltd. Dr Drakoularakou, Professor Rastall and Professor Gibson had no personal or financial conflict of interest.
Financial support was offered through a grant from Clasado Ltd. All data were analysed by Dr Drakoularakou.