Randomised clinical trial: the synbiotic food supplement Probiotical vs. placebo for acute gastroenteritis in children

Authors

  • Y. Vandenplas,

    1. Universitair KinderZiekenhuis Brussel, Vrije Universiteit Brussel, Brussels, Belgium.
    Search for more papers by this author
  • S. G. De Hert,

    1. Department of Anaesthesiology, UZ Gent, Ghent, Belgium.
    Search for more papers by this author
  • PROBIOTICAL-study group

    Search for more papers by this author
    • PROBIOTICAL-group: D. Abrassart, F. Adriaens, N. Balduck, G. Biart, J.-M. Carbonnelle, C. Dauge, P. David, U. Ehrentreich, G. Feron, F. Garbentz, D. Gob, R. Graindorge, B. Hins, A. Hutsebaut, S. Jacquart, I. Jacquemart, C. Leblanc, A.-L. Lenoir, C. Lietaer, K. Logghe, A.-S. Loicq, S. Nouri, J. Poncelet, A. Van Damme, D. Van Damme, L. Van De Vyver, A. Verbist, M. Verboven, I. Verheyden, S. Ballard, Belgium.


Dr Y. Vandenplas, Department of Pediatrics, UZ Brussel, Laarbeeklaan 101, 1090 Brussels, Belgium.
E-mail: yvan.vandenplas@uzbrussel.be

Abstract

Aliment Pharmacol Ther 2011; 34: 862–867

Summary

Background  Some probiotic strains reduce the duration of acute diarrhoea. As a result of strain and product specificity, each product needs support by clinical data.

Aim  In children with acute diarrhoea, to test the efficacy of the synbiotic food supplement Probiotical (Streptoccoccus thermophilus, Lactobacillus rhamnosus, Lactobacillus acidophilus, Bifidobacterium lactis, Bifidobacterium infantis, fructo-oligosaccharides). The primary end-points were duration of diarrhoea and the number of children that had a normalised stool consistency.

Method  A total of 111 children with acute diarrhoea (median age 40 months) were included in this randomised, prospective placebo-controlled parallel clinical trial in primary health care. All children were treated with oral rehydration solution ad libitum and with the synbiotic (n = 57) or placebo (n = 54).

Results  The median duration of diarrhoea was 3 days (IQ 25–75: 2–4 days) in the Probiotical group, compared with 4 days (IQ 25–75: 4–5 days) in the placebo group (P < 0.005). The number of children with normal stool consistency (defined as stool Bristol score ≤4) was higher in the synbiotic group on days 2 and 3 [21 vs. 2% (P < 0.001) and 50 vs. 24% (P < 0.001) respectively]. Less additional medication (antipyretics, antiemetics, antibiotics) was administered in the synbiotic group. Physicians were globally more satisfied with the synbiotic food supplement treatment than with placebo (P = 0.005). One patient in the placebo group was hospitalised.

Conclusion  The median duration of diarrhoea was significantly 1 day shorter in the synbiotic than in the placebo group, associated with decreased prescription of additional medications.

Introduction

Infectious gastroenteritis remains a worldwide problem of health care to date, as there are still approximately four billion diarrhoeal episodes each year. Diarrhoea is defined as an increase in the frequency, water content and volume of stools. Treatment with oral rehydration solution (ORS) is recommended to correct dehydration. ORS has reduced the incidence of mortality and morbidity caused by diarrhoea, but ORS does not shorten the duration of diarrhoea, does not change the consistency of the stools and does not normalise gastrointestinal flora.1

Probiotics are living microorganisms that survive in the gastrointestinal tract and, when ingested in sufficiently large amount, confer a health benefit on the host. Probiotics have been shown to have a luminal action: antitoxic effect against specific pathogens, antimicrobial activity, preserving the tight junctions, modulation of intestinal flora and metabolic activity. Probiotics also have a trophic effect on the mucosa and anti-inflammatory mucosal effects.2

There is evidence that selected strains of probiotics decrease the duration of acute diarrhoea.3, 4 The best studied probiotics include some Lactobacilli species (Lactobacillus (L.) rhamnosus GG, L. reuteri) and Saccharomyces boulardii (S boulardii).3, 4 Many published data focus on populations studied in hospital settings. For some probiotics, such as S. boulardii, the vast majority of data are obtained in ambulatory care and in the developing world.5

As a result of strain and product specificities (especially for products containing mixtures of probiotic strains), and to be in agreement with recommendations of official and scientific organisations, it is recommended to perform randomised controlled trials with each commercialised product. This prospective study was designed to evaluate the additional benefit of Probiotical, a food supplement containing five probiotic strains (Streptoccoccus thermophilus (S. Therm.), L. rhamnosus, L. acidophilus, Bifidobacterium (B.) lactis, B. infantis) and prebiotic fructo-oligosaccharides, above standard ORS treatment on the duration of acute diarrhoea of likely infectious origin in primary health care in Belgian children.

Material and methods

This randomised, prospective, double-blind placebo-controlled trial was conducted in children (age between 3 and 186 months) presented in ambulatory care in Belgium (Table 1: patient characteristics) between April and December, 2010. Acute diarrhoea was defined as the presence of three or more liquid or loose stools, as defined by Bristol criteria ≥type 6 (Table 2)6 per day lasting for less than or equal to 7 days.

Table 1.   Clinical and demographical findings of the study groups
 Probiotical (n = 57)Placebo (n = 54)
  1. Data are expressed as median with range.

Gender (M/F)29/2827/27
Age (months)37 (2–186)43 (1–159)
Duration diarrhoea before inclusion (days)1 (1–7)1 (1–6)
Weight (kg)16.4 (5.9–75.0)17.0 (4.9–38)
Dehydration >5% (%)58
Table 2.   Bristol stool score6
  1. Types 1 and 2 indicate constipation, with 3 and 4 being the ‘ideal stools’ especially the latter, as they are the easiest to defecate, and 5–7 tending towards diarrhoea.

Seven types of stool:
Type 1: Separate hard lumps, like nuts (hard to pass)
Type 2: Sausage-shaped, but lumpy
Type 3: Like a sausage, but with cracks on its surface
Type 4: Like a sausage or snake, smooth and soft
Type 5: Soft blobs with clear cut edges (passed easily)
Type 6: Fluffy pieces with ragged edges, a mushy stool
Type 7: Watery, no solid pieces. Entirely liquid

Inclusion criteria were: an episode of mild to moderate acute diarrhoea (>4 (semi)watery stools/day) according to Bristol criteria (Bristol criteria ≥6)5 of likely infectious origin in infants and children 3 months–15 years for at least 1 day and lasting less than or equal to 7 days. Dehydration was evaluated on clinical grounds and estimated weight loss. Patients had to be mild and moderately dehydrated.2 Exclusion criteria concerned both chronic conditions (such as known chronic uncontrolled intestinal disease such as coeliac disease, cystic fibrosis, food allergy, immune deficiency, inflammatory bowel disease, gastro-intestinal malformations, abnormal gastrointestinal motility and more acute conditions (antibiotic treatment during the preceding 7 days, the presence of macroscopic blood in the faeces and use of probiotic products except for the probiotics present in infant formula, if the patient developed the diarrhoea while being fed with this formula). The primary endpoint was the duration of diarrhoea and the number of children who had a normalised stool consistency according to the Bristol criteria (stool score ≤4) during the study. Normalisation of stool consistency is equivalent to the duration of diarrhoea. The secondary endpoint was the number of stools per day.

All patients were treated according to the recommendations with the same hypo-osmolar oral rehydration solution (ORS) ad libitum (Soparyx; Na+ 60, Cl 50, K + 20, citrate 10 mmol/L; osmolarity 140 mmol/L).2, 3 Physicians were allowed to prescribe additional medication according to what they would consider ‘good clinical practice’ if the response of the patient to the study-treatment (ORS and synbiotic or placebo) was not satisfactory.

Every participating physician was active in primary health care and received 10 blinded treatments. The tested synbiotic was administered as a capsule that could be opened and mixed in ORS or water. The placebo was the same capsule without the probiotics. Both products were supplied by Phacobel (Tinlot, Belgium), who had no role in the concept, design or conduct of the study or in the analysis or interpretation of the data. The active product and placebo were packed in identical boxes; they were of the same colour, weight, smell and taste. Both the physicians and the patients were unaware of the real nature of the product. The unblinding procedure was performed after the study was completed and after the statistical analyses were finalised. Randomisation for every participating physician was done by computer. Patients were enrolled according to the computer-determined allocation to product A or B. Random allocation was made in blocks of ten to obtain groups of similar size. The sequence was concealed until treatments were assigned. Participating physicians had to open a neutral envelope to know to which group the patient was allocated. Group A received the active product (Probiotical, Laboratoires Phacobel Belgium, Table 3), and group B received placebo – 1 capsule/day for 7 days. All patients were examined by the treating physician at inclusion and at day 4. The physicians registered demographical data, degree of dehydration and duration of diarrhoea before inclusion, and if hospitalisation occurred. Before inclusion of the patient, all inclusion and exclusion criteria had to be registered. If a patient did not fulfil the inclusion criteria, participation in the study trial was not possible. Parents recorded in a specific diary the number and the aspect of the stools according to the Bristol criteria (illustrations were provided), and the use of any other medication. Day one was defined as ‘the first 24 h of treatment’. As this trial was performed in primary health care, blood sampling or stool analyses were not performed.

Table 3.   Composition of Probiotical
  1. Cost: 6.8 €/10 capsules. Available in Belgium and Luxemburg; or throughout Europe if ordered by pharmacists.

Streptococci6.5 × 109Streptococcus thermophilus60 mg
Lactobacilli6.5 × 109Lactobacillus rhamnosus28 mg
Lactobacillus acidophilus28 mg
Bifidobacteria6.5 × 109Bifidobacterium infantis20 mg
Bifidobacterium lactis20 mg
Fructo-oligosaccharides  20 mg
Ascorbic acid  1.2 mg

This trial was not registered in a publicly accessible registry. The study was approved by the institutional ethical committee, and parents gave their written consent. To obtain the required power (95%, type 1 error = 0.05, two tailed test), we needed 45 participants in each group. This estimate assumes a mean difference in duration of diarrhoea of 1 day (24 h) between the treated and control children (corresponding to means of 120 and 96 h) with a s.d. of 30 h within the group. This assumption is based on the results of similar trials. Statistical analysis was performed using the sigmastat 2.03 software package (SPSS, Leuven, Belgium). Variables were tested for normal distribution using the Kolmogorov–Smirnov test and compared using the Mann–Whitney U-test and Chi-square or Fisher’s exact tests, as appropriate. Only an ‘intention-to-treat’ analysis was performed. Statistical significance was accepted at P < 0.05. The following factors – gender, age, weight, duration of diarrhoea before treatment and group assignment – were entered in a backward stepwise regression analysis to test for independent contributions of the different risk factors to the duration of diarrhoea.

Results

The study was proposed to the parents of 140 eligible patients, presenting with acute diarrhoea (median duration 1 day; range 0–7 days) as major symptom of an acute gastroenteritis likely of infectious origin. Twenty-nine parents refused participation because of the risk of receiving placebo; 111 children were included (Table 1). Fifty-seven patients were allocated to Group A (Probiotical group) and 54 patients to group B (placebo group). Demographical characteristics and nutritional status were comparable. At inclusion, the degree of dehydration in both groups was similar (Table 1).

No patient was lost in follow-up before stool consistency had normalised; no patient was excluded from the analysis. The median duration of diarrhoea (Bristol score ≤4) was 1 day shorter (95% confidence interval −1.9–−0.6 days). Median duration of diarrhoea was 3 days (range 1–6 days; IQ 25–75: 2–4 days) in the Probiotical group and 4 days (range 2–6 days; IQ 25–75: 4–5 days) in the placebo group (P < 0.005). The 95% confidence intervals for the duration of diarrhoea were 3–4 days and 4–5 days respectively (= 0.005). The relative risk of having diarrhoea for less than 4 days with Probiotical compared to placebo was 0.44 (95% confidence intervals: 0.26–0.75; P = 0.0018). Backward stepwise regression analysis identified group assignment as the only independent factor determining duration of diarrhoea (P < 0.001). None of the other factors significantly influenced the duration of diarrhoea.

One patient of the control group was hospitalised. The number of children with a normalised stool consistency (Bristol scale ≤4) was significantly higher on day 2 and day 3 (Table 4) in the Probiotical group. On day 2, there was a reduction in the number of stools in the active treatment group compared with the placebo group (Table 5). In the placebo group, 38/54 (70%) patients received additional medication (mainly antipyretics, antiemetics and antibiotics), whereas in the Probiotical only 11/57 (19%) patients received additional medication (P < 0.0001). Side effects were not reported by the parents or physicians. Assessment of the satisfaction of treatment by the physicians was scored as acceptable or better in both treatment groups, but globally, physicians were more satisfied with the synbiotic food supplement than with placebo (P = 0.005) (Table 6).

Table 4.   Percentage of children with normal stool (Bristol score ≤4)
 TreatmentPlaceboP
Day 11/57 (2%)0/54 (0%)1.000
Day 212/57 (21%)1/54 (2%)0.002
Day 325/51 (50%)13/53 (24%)0.014
Day 432/36 (89%)37/40 (92%)0.702
Table 5.   Number of stools/day in both groups
 ProbioticalPlaceboP
MedianRangeIQ 25–75MedianRangeIQ 25–75
  1. Data are expressed as median (range) (Interquartiles 25–75).

Day 151–114–651–84–6N.S.
Day 231–82–440–82–50.04
Day 320–61–321–62–3N.S.
Day 421–51–320–41–3N.S.
Table 6.   Treatment satisfaction as estimated by the physicians on day 4
 TreatmentPlacebo
  1. Outcome of treatment was always assessed as at least acceptable by the physicians. Global satisfaction was better in the treatment group than in the placebo group (P = 0.005).

Poor0 (0%)0 (0%)
Acceptable0 (0%)4 (7%)
Good8 (14%)15 (28%)
Very Good24 (44%)25 (48%)
Excellent23 (42%)9 (17%)

Discussion

In Europe, probiotics are advised besides appropriate rehydration and patient education in the management of acute diarrhoea in childhood.3 Probiotics are live microorganisms that survive in the gastrointestinal tract and have a health benefit on the host. Some probiotic strains have been shown to offer a safe intervention for acute infectious diarrhoea and reduce the duration and the severity of the disorder. A recent Cochrane review included 63 trials (of whom 56 in children) with 8014 participants and concluded that the probiotics tested reduced duration of diarrhoea within 24.76 h (95% CI 15.9–33.6 h; n = 4555, trials = 35).4 Also, according to the same Cochrane review, the frequency of stools was decreased on day 2 (mean difference 0.80; 0.45–1.14; n = 2751, trials = 20).4 The results of the synbiotic tested in this trial (Probiotical) are comparable or even moderately better. Both on day 2 and 3 stool consistency had normalised in significantly more patients using the active product compared with the placebo group. Results may be dose and age dependent.7, 8

Mixtures of probiotics have been shown to have beneficial effects on different end points including diarrhoea, irritable bowel syndrome, gut function, atopic disease, immune function and respiratory tract infections, gut microbiota modulation, inflammatory bowel disease and treatment of Helicobacter pylori infection.9 However, only 16 studies compared the effect of a mixture with that of its component strains separately, although in 12 cases (75%), the mixture was more effective.9 In some studies, one strain is equally effective or even slightly more effective than a mixture (S. boulardii, L. acidophilus, L. rhamnosus, B. longum) with different probiotics including the single strain (S. boulardii).10 The synbiotic product tested (Probiotical) contains several microorganisms, for which there is evidence of efficacy. Micro-organisms for which previous publications have suggested efficacy in reducing the severity and duration of acute diarrhoea in children include strains present in Probiotical, such as L. rhamnosus (formerly ‘L. casei strain GG’ or ‘L. GG’). L. GG is one of the best studied bacterial probiotics.2, 3 According to the results of the study by Canani and co-workers, L. Rhamnosus (strain GG) and a mix of four bacterial strains (L. Delbrueckii var bulgaricus, S. Therm., L. Acidophilus and B. Bifidum) were effective in reducing the duration of diarrhoea.11 A formula with Str. Therm. and B bifidum, two of the four bacterial species in the preparation studied, protected chronically sick children aged below 24 months against diarrhoea (prevention).12

Synergism between different strains has been demonstrated. Juntunen et al. showed a better adherence of B. Bb12 in presence of L. rhamnosus GG.13 The better the adhesion occurs, the more IgA secretion.13 Whether the greater efficacy of multi-strain products compared with single strain products9 is due to synergistic interactions between strains or a consequence of the higher probiotic dose used remains unclear.

To date, few studies evaluated the efficacy of synbiotics, a mixture of probiotics and prebiotics in the treatment of infectious diarrhoea. The product tested, Probiotical, is a synbiotic, as it contains a very small amount of prebiotic oligosaccharides (20 mg of oligofructose). For comparison, the amount of oligosaccharides in starter infant formula is about 800 mg per 100 mL or 8 g, if a baby drinks 1 L of formula. In other words, it can be questioned if the tiny amount of 20 mg present in a capsule of Probiotical can have a relevant clinical effect. Nondigestible carbohydrates were shown not to reduce the duration of diarrhoea,14 although a synbiotic may confer additional benefits over a probiotic by increasing bifidobacteria levels.15

Stool cultures were not performed, as is the case in many studies with probiotics.14, 16 According to the evidence-based guidelines of the European Society of Paediatric Gastroenterology, Hepatology and Nutrition and the European Society of Infectious Disease, it is not recommended to perform stool cultures for acute gastroenteritis in primary health care.3

This is the first study in which physicians were asked for their satisfaction with treatment outcome. It was striking to observe that, although there was overall no difference between both groups regarding ‘satisfaction’ with treatment outcome, over 40% of the physicians did evaluate the outcome as ‘excellent’ in the treatment group, whereas this was only 17% in the case of control group (Table 6). This is also the first time that a decreased administration of additional medications such as antipyretics, antiemetics and antibiotics is documented.

In conclusion, the synbiotic Probiotical reduces the duration of diarrhoea of likely infectious origin to at least a similar extent as most data published, because stool consistency normalised faster, resulting in a reduction of the duration of diarrhoea in the treatment group compared with the control group. The addition of a prebiotic may induce a theoretical benefit to normalise the gastro-intestinal flora composition of the host more rapidly (although the amount of fructo-oligosaccharides is small). The number of stools was significantly reduced on day 2. Physicians did evaluate outcome better in the treatment group. The reduced prescription of additional medication illustrates the better outcome and results in an additional economic benefit.

Acknowledgements

Declaration of personal interests: Yvan Vandenplas has served as a speaker, consultant and advisory board member for Abbott, AstraZeneca, Biocodex, Biogaia, Danone (Nutricia, SHS), Mead Johnson, Nestle Nutrition and United Pharmaceuticals. Declaration of funding interests: The study was funded (only providing study materials) by Phacobel Belgium. No funding was received for writing or preparation of the manuscript.

Ancillary