SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

Background

: Prebiotic carbohydrates selectively stimulate the growth of bifidobacteria and lactobacilli in the human colon. These bacteria form part of the gut’s inherent defence against invading pathogens.

Aim

: To test the effectiveness of fructo oligosaccharides in preventing travellers’ diarrhoea.

Methods

: A total of 244 healthy subjects, travelling to high and medium risk destinations for travellers’ diarrhoea, took part in a randomized, double-blind, placebo-controlled study. The protocol comprised a preliminary week for recording bowel habit by diary, a 2-week pre-holiday period with the diary and consumption of 10 g of fructo oligosaccharides or placebo daily, followed by a 2-week holiday with continuation of treatment and diary. A post-study questionnaire was completed by all subjects on their return to the UK.

Results

: The consumption of fructo oligosaccharides led to a small (6%; P < 0.02) increase in stool frequency in the pre-holiday period and gave a significantly better sense of ‘well-being’ during the holiday, although subjects reported more flatulence. There were non-significant decreases in episodes of diarrhoea with 20% on placebo and 11% on fructo oligosaccharides recording episodes in the post-study questionnaire (P=0.08) and 46% placebo, 38% fructo oligosaccharides recording episodes in the diary (P > 0.1). No change in bowel frequency, consistency or stool size was recorded.

Conclusion

: Travel to high risk areas increases diarrhoea. Fructo oligosaccharides alone are not sufficient to prevent this, although do have some benefits for the subjects.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

Travellers’ diarrhoea remains a common problem for those journeying overseas for either business or holidays. It is particularly frequent in visitors to Central America, the Far East, India and parts of Africa. Current estimates are that 60 million travellers from the West visit high risk areas annually, and of these 30–50% have episodes of diarrhoea.1[2][3][4][5][6]–7

The infecting organisms are very much the same as those that cause acute diarrhoea in both developed and developing countries, e.g. salmonella, shigella, campylobacter, enterotoxigenic and other E. coli, protozoa such as Giardia lamblia and Entamoeba histolytica, and viruses, especially rotavirus.5, 8[9][10]–11 Of the organisms responsible for travellers’ diarrhoea, enterotoxigenic E. coli is by far the most common and is isolated in around 40% of cases. Salmonella and shigella account for a further 25%, followed by giardia and amoeba 15%; the rest are campylobacter, viruses, etc. However, in 20–40% of cases no pathogen is detected. Therefore, of people who get travellers’ diarrhoea, around 20% can probably be said to have a cause which is specifically targeted at the large intestine.1, 3, 10, 11

Gastrointestinal infection requires not only an initiating dose of organisms, but also a suitable environment in the gut in which to grow. Part of the defence against invading organisms is the indigenous flora of the hind gut which, through a variety of mechanisms, prevent the establishment of newly arrived species. Amongst the bacteria which are said to be important in this context are bifidobacteria and lactobacilli, both of which secrete bacteriostatic peptides, which are part of the process whereby invading pathogens are repelled.12[13]–14

A preventive and therapeutic strategy which has been tried in the past is to supplement those at risk of diarrhoea, or suffering an acute attack, with oral doses of probiotic organisms, such as lactobacilli.15, 16 These bacteria have been shown to survive gastric acid and adhere to small intestinal mucosa. Through this mechanism they are thought to prevent the adhesion of pathogenic bacteria and thus protect against diarrhoea or reduce the severity of attacks. However, in published studies in which probiotic bacteria have been tried in travellers’ diarrhoea, the degree of protection has been relatively small, at around 10%.17[18][19]–20

In previous studies, Gibson and colleagues have shown that, in vitro, when bifidobacteria are given fructo oligosaccharides as a substrate, they secrete a peptide that is inhibitory to most of the common organisms causing acute diarrhoea.13 In human healthy volunteer studies, oral dosing with 15 g of fructo oligosaccharides daily led to bifidobacteria becoming the dominant species in the large bowel.21

Therefore, the hypothesis to be tested is that altering the environment in the hind gut towards a flora in which bifidobacteria are the predominant species leads to a greater resistance to invading pathogens. This alteration can be achieved with oral fructo oligosaccharides, which may therefore provide a simple way of reducing the morbidity in travellers to endemic regions of diarrhoea.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

The study was a placebo-controlled, randomized, double-blind parallel design lasting 5 weeks for each subject.

Subjects and recruitment

Healthy volunteers were recruited through advertisement in the consumer press. Respondents to the advert were contacted by telephone and the study was explained to them. They were then asked a series of screening questions about their holiday, and health and demographic data were obtained. Details of the reasons for including or excluding potential volunteers are given in Table 1. If suitable, subjects were then sent a volunteer information form and a consent form; their General Practitioners were contacted by letter and given an explanation of the study. The GPs were asked to complete a questionnaire about their patients and, particularly, to say whether there were any factors that should lead to their exclusion from the study. Three GPs said that their patients were unsuitable to take part. Volunteers had to be travelling to a high risk destination for travellers’ diarrhoea, as defined in the guide produced by the Department of Travel Medicine, The Hospital for Tropical Diseases, London. Seventy-eight percent of the volunteers went to high risk destinations and 22% to medium risk (Table 2). Details of the 244 volunteers who started the study are given in Table 3. Four subjects completed the study but were excluded from the results because their diaries were incomplete.

Table 1.   Inclusion/exclusion criteria for travellers’ diarrhoea study Thumbnail image of
Table 2.   Destination of 240 subjects who completed the travellers’ diarrhoea study Thumbnail image of
Table 3.   Baseline characteristics of the placebo and fructo oligosaccharides groups Thumbnail image of

Protocol

After recruitment, volunteers were given a study number and randomly assigned by computer to either fructo oligosaccharides or placebo in blocks of 20, stratified for sex. Volunteers were then sent a pack of materials containing:

• 60 sachets of the test material (placebo or fructo oligosaccharides);

• three diaries and instructions;

• an explanatory letter for Customs and Excise;

• a travellers’ diarrhoea self-treatment guide; and

• a volunteer information sheet (for reference).

Three weeks before departure, volunteers were asked to fill in a diary for 1 week to detail their usual bowel habit. The diary was one which has been used for many years at the Dunn Clinical Nutrition Centre and comprises a 14-day diary with a page for each day. The subjects were required to record the amount and consistency of stools, whether they had any discomfort and to tick those symptoms they had experienced from a list that included: fever; abdominal pain; vomiting; abnormal bloating; chronic flatulence; diarrhoea; and constipation.

After 1 week of baseline recording, subjects were asked to commence taking the test material with the recommendation to take one sachet in the morning and one in the evening, dissolved in water. Each sachet contained either 5 g of fructo oligosaccharides or 5 g maltodextrin. At this time subjects were also asked to begin keeping a new 2-week pre-holiday diary. On holiday, subjects continued with the test material and completed a further 2-week diary.

On return from holiday, subjects were sent a post-study questionnaire asking 12 questions about compliance with the regime, experience of diarrhoea and other symptoms, general feeling of well-being, medication taken, etc. Subjects were asked to mail the post-study questionnaire and bowel habit diaries back to the study team and on receipt of these were sent £25 as a gift for completing the study.

Outcome measures

Three primary outcome measures for the study were used:

1 The total number of stools over the 2 weeks of the holiday.

2 The sum of the size scores for each stool (scored as 1=small, 2=medium, 3=large) over the 2 holiday weeks.

3 The sum of the consistency scores for each stool (scored as 1=hard, 2=soft, 3=loose) over the 2 holiday weeks.

The latter two outcome measures were adjusted for the first, effectively giving mean scores per stool size and consistency, respectively.

In addition, daily symptoms of bloating, flatulence, diarrhoea and constipation were scored as present or absent, and these scores were summed over the three periods of the study: baseline, pre-holiday and holiday, and analysed in the same way as for stool frequency, size and consistency.

Statistics

For the power calculation, we predicted that about 50% of the subjects would get diarrhoea and that fructo oligosaccharides would reduce this attack rate by 20%. For 80% power and 5% significance this gave 93 subjects per treatment group. Additional subjects were recruited to allow for those not completing the study satisfactorily. The final numbers to start the study were 127 subjects taking placebo and 117 subjects taking fructo oligosaccharides.

Data were analysed by χ2-test, Fisher’s exact text, t-test and multiple linear and logistic regression analysis. In the regression analysis the dependent variable was one of the outcome measures above, and independent variables were included, corresponding to the outcome measure (i.e. number, size or consistency) for the baseline and pre-holiday periods. This means that each subject’s bowel score on holiday was adjusted for his or her corresponding pre-holiday and baseline scores. Other independent variables were included where appropriate to adjust for other relevant baseline characteristics, e.g. age, sex, travellers’ diarrhoea (TD) risk of holiday destination (high vs. medium), recent travel abroad and recent TD.

Finally, a dummy variable indicating the subject’s randomization was included, to test for a treatment effect after adjusting for all relevant confounders. The two treatments were coded blind as 8398 and 8297, and the effect of 8297 (active) relative to 8398 (placebo) was coded as a variable named ‘8297’.

Owing to skewness in the distributions, the bowel habit summary statistics were all transformed to natural logarithms and multiplied by 100 for the regression analysis. This gives the results in percentage units.22

Ethics

Ethical approval for the study was granted by the MRC Dunn Clinical Nutrition Centre Ethical Committee in April 1998. The first subjects were recruited on 30th April 1998 and the last on 29th September 1998.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

A total of 363 subjects were recruited and randomized, of whom 119 subsequently dropped out. Overall, there were more drop-outs in the fructo oligosaccharides group (64) compared to the placebo group (55), but there was no difference in the male : female ratio or holiday destination. Table 4 gives the reasons for 119 subjects not starting the study; the principal reasons for not starting were holiday cancellations and pre-study illness. The adverse reactions reported by four subjects were flatulence.

Table 4.   Reasons why 119 volunteers did not start the study (n = 115) or dropped out during the pre-holiday period (n = 4) Thumbnail image of

The principal destinations and numbers of volunteers who visited each place are given in Table 2. The distribution of destinations was similar between the placebo and fructo oligosaccharides groups.

In total, 244 subjects completed the study, of which 117 were fructo oligosaccharides and 127 placebo. Table 3 gives baseline data for each group.

Table 5 summarizes the bowel habit data for the three time periods in the two groups. One subject did not complete the pre-holiday diary and three subjects had incomplete diaries. There were no significant differences in bowel habit between the groups at any point, as indicated by the diaries.

Table 5.   Bowel habit by study period and treatment Thumbnail image of

Regression analysis relating the change in stool frequency due to fructo oligosaccharides in the pre-holiday period adjusted for baseline stool frequency, showed a significant (6%; P=0.02) increase in frequency with fructo oligosaccharides. Similar analyses for stool size and consistency did not show an effect. During the holiday period, bowel frequency was significantly related to baseline pre-holiday frequency, indicating substantial heterogeneity amongst the group of subjects. Women showed 7% less frequency than men (P=0.03) and there was a small trend to increased frequency with age (P=0.001). Overall, stool size, consistency and frequency were all significantly and positively related.

Bowel habit and symptoms

To relate bowel habit to clinical symptoms, regardless of treatment, regression analysis was carried out against reports of fever, abdominal pain, vomiting, bloating, flatulence, diarrhoea, constipation, improved regularity, irritable bowel symptoms and general well-being, from replies in the post-study questionnaire for the holiday period. Adjustments were made for baseline and pre-holiday scores. Stool frequency was significantly increased (10.2%; P < 0.02) for diarrhoea and for improved regularity (7.8%; P < 0.05). This is to be expected and provides some validation of the bowel habit diary. Similarly, stool size was significantly reduced in those complaining of constipation, by 9.1% (P < 0.006), and stool consistency was looser when diarrhoea was a complaint, by 10.8% (P < 0.002) and harder in constipation, by 16.6% (P < 0.001), with associated bloating.

The post-study questionnaire also recorded side-effects from the two treatments and the results are given in Table 6. When compared to placebo, fructo oligosaccharides caused significantly more flatulence (18.8% vs. 34.5%), which is to expected, but also a greater feeling of well-being (4.7% vs. 12.9%). Reported attacks of diarrhoea were also less with fructo oligosaccharides (19.5% vs. 11.2% of persons), but this difference was not statistically significant (P=0.08).

Table 6.   Frequencies (%) of side-effects, beneficial and negative, of the two treatments as recorded in the post-study questionnaire Thumbnail image of

The daily diary, in addition to recording stool frequency, size and consistency, also gave the subjects a chance each day to record any symptoms they had experienced by ticking either fever, abdominal pain, vomiting, bloating, flatulence, diarrhoea and constipation. Consistent with the results from the post-study questionnaire, there was less diarrhoea reported by subjects on fructo oligosaccharides when on holiday (45.7% vs. 38.2%), but again the results were not significant (χ2=1.39).

A summary of symptoms recorded in the daily diaries is shown in Table 7 for all subjects regardless of treatment. A striking increase in gastrointestinal symptoms is seen during the holiday period. Episodes of diarrhoea were reported by only 8% of subjects in the pre-holiday period, but by 40% during the holiday. Similarly, constipation increased from 12 to 23% and episodes of abdominal pain from 5 to 29%. There was no change in reporting of bloating and flatulence.

Table 7.   Subjects reporting symptoms in diaries (n = 240) Thumbnail image of

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

Although standards of hygiene have improved substantially in many countries over the years, nevertheless, episodes of diarrhoea and other gut-related symptoms remain a troublesome problem for travellers and can ruin a much anticipated holiday or important business trip. Strategies for the prevention of travellers’ diarrhoea in the past have included the avoidance of local water supplies, salads and unpeeled fruit, reheated food and any dish containing meat or meat products. On occasions, antibiotic prophylaxis has been recommended. Despite this, up to 50% of travellers to high risk areas experience diarrhoea and new preventive measures are being sought. One possibility that has emerged recently is the use of probiotic bacteria, which have been shown to ameliorate attacks of rotavirus, diarrhoea in children and to be equally as effective as conventional therapies in treating inflammatory bowel disease.15, 17[18][19]–20, 23[24][25]–26 However, the benefits with probiotics in travellers’ diarrhoea have so far proved to be small, and better solutions are needed.

This study presents, for the first time, the use of a pre-biotic, fructo oligosaccharides, in the prophylaxis of travellers’ diarrhoea. Prebiotics selectively stimulate the indigenous flora that normally strengthen the barrier to invading pathogens, namely, lactobacilli and bifidobacteria. If the natural colonization resistance which is present in the gut to invading organisms is strengthened, so too will be the resistance to travellers’ diarrhoea.

The results of the present study do not provide convincing evidence that fructo oligosaccharides on their own will prevent all causes of travellers’ diarrhoea. Nevertheless, the study provides a number of important observations which can be used to design future trials.

From the post-study questionnaire responses, a clear effect of the fructo oligosaccharides was to increase flatulence in subjects. This is not unexpected because flatulence is the main complaint in most published studies aimed at determining the physiological and metabolic properties of these prebiotic sugars.27 The flatulence was not sufficiently severe to discourage more than three subjects from continuing with the study and is a good indicator that compliance with the regime was high. Unexpectedly, subjects taking fructo oligosaccharides noticed a significant improvement in well-being during their holiday. There is no immediately apparent reason for this, although the fate of fructo oligosaccharides in the large intestine is to be fermented. The resultant short chain fatty acid production and alterations in microbial metabolism may possibly provide the basis for an explanation.

In both the diary and post-study questionnaire, subjects on fructo oligosaccharides reported reduced episodes of diarrhoea, although in neither case was the change statistically significant. The more objective measures of bowel habit, stool frequency, consistency and size, recorded in the diary, showed no differences between the two groups and a diarrhoea severity score constructed from the number of loose stools over the holiday period, together with the total number of episodes of fever, abdominal pain and flatulence also showed no difference between the two groups.

A possible reason for the failure of the study to demonstrate a clear effect, could lie in the study design. The subjects were recruited through consumer magazines, probably read by mainly health conscious individuals of whom the majority were over 50 years of age and who might already be taking all the necessary precautions to avoid travellers’ diarrhoea. However, the prevalence of diarrhoea in this group was 42% and as such, reflects very closely other reported incidence rates in widely differing populations.1[2][3][4][5][6]–7 Compliance with the regime was also good. Dropouts all occurred before travel and no sachets were returned after the holiday.

A more likely reason for the failure of fructo oligosaccharides to prevent travellers’ diarrhoea may lie in its mechanism of action and in the aetiology of diarrhoea in travellers. Fructo oligosaccharides primarily affect the large intestinal microflora and may well improve colonization resistance. However, there are multiple causes of diarrhoea in travellers other than those which target the large intestine. Many pathogens affect primarily the small intestine; for example, enteropathogenic E. coli, campylobacter, giardia and yersinia. In order to prevent the effect of these and similar organisms, a preventive strategy aimed at the small intestine would be required. Furthermore, in all reported studies where the aetiological agent of diarrhoea has been sought in travellers, in at least 20%, no organism has been identified.10, 11 Whilst this failure to identify the incriminating pathogen has often been ascribed to the inadequacy of microbiological methods, sampling problems, etc., it is equally likely that subjects experience episodes of diarrhoea for reasons other than simple infection. Many people travelling abroad eat strange foods to which they may never have been exposed and which may cause food intolerance and diarrhoea. Equally, people on holiday may consume far more alcohol than usual and again this is known to upset bowel habit. Overall therefore, the multiple causes of diarrhoea in travellers make it unlikely that a single preventive strategy will be effective.

One striking finding from this study is the number of people who have some change in bowel habit whilst travelling (Table 7). Apart from the 40% who experienced diarrhoea, others reported constipation; 29% had one or more episodes of abdominal pain and 35% reported flatulence and bloating. Overall, 69% of subjects reported some disturbance in bowel function whilst on holiday.

In the future, it may well be possible to reduce the burden of infectious diarrhoea and possibly other disturbances of bowel habit in travellers by combining a pro and prebiotic mixture. Probiotic organisms probably target the small intestine through their properties to adhere to mucosal surfaces and therefore either displace, compete with or exclude invading pathogenic species. This is presumably the mechanism seen in the prevention of infectious diarrhoea in children.23, 24 Prebiotics which act mainly in the large intestine would therefore act in synergy with probiotics and may, together, produce an effective remedy.

ACKNOWLEDGEMENT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References

This study was supported by Lambert Healthcare.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. References