Background
Definition
Diarrhoea is defined by the World Health Organization (WHO) as 3 or more loose or watery stools (taking the shape of the container) in a 24-hour period. Diarrhoea is acute if the illness started under 14 days previously, and persistent if the episode has lasted 14 days or more (Anonymous 1988). Normal infants who are exclusively breastfed may pass loose, "pasty" stools frequently. In this group, the definition is usually based on what the mother considers to be diarrhoea (WHO 1990). Infectious diarrhoea is an episode of diarrhoea that is caused by an infectious agent.
Incidence and mortality
Infectious diarrhoea occurs much more commonly in developing countries than industrialized countries (Guerrant 1990). Attack rates in developing countries are typically 6 to 12 episodes per child per year, compared with 2 in the USA (Savarino 1993). In developing countries, deaths are most common in children younger than 5 years, and account for 2.4 to 3.3 million deaths each year (Bern 1992). In industrialized countries, deaths are mainly in the elderly (Savarino 1993).
Causes
Many infectious agents cause diarrhoea. Worldwide, rotavirus is the most common cause of severe diarrhoea and diarrhoea mortality in children (Cunliffe 1998). Other important viral pathogens are adenoviruses and enteroviruses. Important bacterial pathogens are: enterotoxigenic Escherichia coli, Salmonella, Shigella, Yersinia, Campylobacter, and Vibrio cholera. The main parasitic causes of diarrhoea are Cryptosporidium and Giardia. An aetiological study of young children attending hospitals in China, India, Mexico, Myanmar, and Pakistan showed that rotavirus, enterotoxigenic E. coli and Shigella spp. were the most commonly isolated pathogens (Huilan 1991). Acute diarrhoea is frequent among travellers in whom enterotoxigenic E. coli is particularly common (Black 1986). In practice, most episodes of acute diarrhoea that are assumed to be caused by an infectious agent are treated without the causative agent being identified. Major causes of acute infectious diarrhoea will differ according to local factors such as availability of clean water and sanitation. In contrast with acute infectious diarrhoea, infection is likely to be only one of several factors that contribute to the pathogenesis of persistent diarrhoea (Walker-Smith 1993).
Treatment
The aim of treatment is to prevent or reverse dehydration, shorten the length of the illness, and to reduce the period that a person is infectious. Treatment options available are oral rehydration solution, antibiotics, and gut motility suppressing agents such as loperamide, codeine, and probiotics. This review considers the use of probiotics only.
Probiotics
Probiotics have been defined as microbial cell preparations or components of microbial cells that have a beneficial effect on the health and well being of the host (Salminen 1999). Fermenting foods to enhance their taste and nutritional value is an ancient and widespread practice. Well-known probiotics are the lactic acid bacteria and the yeast Saccharomyces (Naidu 1999). The taxonomy of the lactic acid bacteria relied traditionally on phenotypic characteristics. Modern molecular techniques have shown these to be unreliable, and "polyphasic taxonomy" using both phenotypic and molecular techniques is now recommended (Klein 1998). Even closely related probiotic strains can have different clinical effects, and the Food and Agricultural Organization of the United Nations (FAO) and WHO expert consultation committee emphasized that beneficial effects observed with one strain cannot be assumed to occur with other strains (FAO/WHO 2001). This implies that reliable identification of organisms at the strain level is necessary for clinical studies.
The rationale for using probiotics in infectious diarrhoea is that they act against enteric pathogens by competing for available nutrients and binding sites, making the gut contents acid, producing a variety of chemicals, and increasing specific and non-specific immune responses (Gismondo 1999; Goldin 1998; Vanderhoof 1998). No serious adverse effects of probiotics have been suggested in well people, but infections have been reported in people with impaired immune systems (Hata 1988; Piarroux 1999; Salminen 1998; Saxelin 1996; Sussman 1986).
Two systematic reviews of probiotics in acute diarrhoea have been published. Szajewska 2001 included only published, randomized, placebo-controlled, double-blind studies of acute diarrhoea lasting 3 or more days in infants and children. A score was used to assess methodological quality. The effects of all probiotics and of individual strains were analysed. The risk of diarrhoea lasting 3 or more days was reduced by 0.40 in the probiotic compared with the placebo group (95% confidence interval (CI) 0.28 to 0.57, random effects model; 8 trials including 731 children) and probiotics reduced the duration of diarrhoea by 18.2 hours (95% CI 9.5 to 26.9 hours, random effects model; 8 trials including 773 children). Statistically significant heterogeneity in this result was resolved when one study, which employed a mixture of three probiotic organisms, was excluded. Lactobacillus GG was thought to be particularly effective in rotavirus diarrhoea.
A meta-analysis undertaken by Van Niel 2002 was restricted to adequately randomized and blinded studies of several strains of lactobacilli in children. Children who had received recent antibiotics were excluded. Probiotics reduced the duration of diarrhoea by 0.7 days (95% CI 0.3 to 1.2 days; 7 studies including 675 children) and diarrhoea frequency on day 2 by 1.6 (95% CI 0.7 to 2.6; 3 studies including 122 children). Heterogeneity of results between studies prevented the analysis of the effects of individual strains of lactobacilli.
Our review aims to maximize use of available data by including participants of all ages, unpublished studies, and non-blinded ('open') studies. Also, rather than using a score, we assessed the relevant methodological aspects of trials individually (Juni 1999). These were the generation of allocation sequence, allocation concealment, blinding, and loss to follow up. To maximize the relevance of our findings for clinical practice, we included studies in which participants with infectious diarrhoea had received antibiotics prior to recruitment.
Objectives
To assess the effects of probiotics in proven or presumed infectious diarrhoea.
Methods
Criteria for considering studies for this review
Types of studies
Randomized controlled trials.
Types of participants
Adults and children with acute diarrhoea (duration <14 days) that is proven or presumed to be caused by an infectious agent.
Excluded: studies of diarrhoea known or thought to have other causes (eg antibiotic-induced diarrhoea) and studies of persistent diarrhoea.
Types of interventions
Intervention
Specific, identified probiotic.
Excluded: yoghurt or other fermented foods in which a specific probiotic agent was not identified.
Control
Placebo or no probiotic.
Intervention and control arm to be otherwise treated identically in relation to other treatments and drugs.
Types of outcome measures
Primary
- Diarrhoea lasting 3+ and 4+ days.
- Duration of diarrhoea.
- Stool frequency and volume.
Secondary
- Need for unscheduled intravenous rehydration after randomization.
- Deaths.
- Adherence.
- Adverse events, such as vomiting.
- Withdrawal from trial.
Search methods for identification of studies
We have attempted to identify all relevant studies regardless of language or publication status (published, unpublished, in press, and in progress).
We searched the Cochrane Infectious Diseases Group's trials register in December 2002 using the search terms: diarrhea/; diarr$(tw); diarhea(tw); probiotic(tw); Lactobacill$(tw); Lactococc$(tw); Bifidobacter$(tw); Enterococc$(tw); Streptococc$(tw); Saccharomyces(tw). Full details of the CIDG methods and the journals handsearched are published in The Cochrane Library in the section on 'Collaborative Review Groups'.
We searched the Cochrane Controlled Trials Register published on The Cochrane Library (Issue 4, 2002) using the search terms: diarrhea/; diarr$(tw); diarhea(tw); probiotic(tw); Lactobacill$(tw); Lactococc$(tw); Bifidobacter$(tw); Enterococc$(tw); Streptococc$(tw); Saccharomyces(tw).
We searched MEDLINE (1966 to 2002) and EMBASE (1988 to 2002) using the search strategy defined by The Cochrane Collaboration (Clarke 2003) and following search terms: diarrhea/; diarr$(tw); diarhea(tw); probiotic(tw); Lactobacill$(tw); Lactococc$(tw); Bifidobacter$(tw); Enterococc$(tw); Streptococc$(tw); Saccharomyces(tw).
We contacted organizations and individuals working in the field, and the following pharmaceutical companies that manufacture probiotic agents to help identify additional published trials and unpublished data: Biogaia Biologicals, Lund, Sweden; Nestle Foundation, Lausanne, Switzerland; Probiotics International Ltd, Somerset, UK; Ross Products Division of Abbott Laboratories, Ohio, USA; and Yakult, London, UK.
We also drew on existing reviews of this topic and checked the citations of all trials identified by the above methods.
Data collection and analysis
Study selection
Stephen Allen (SD) and Leonila Dans (LD) independently reviewed the titles of papers and, where available, abstracts generated by the search to identify potentially relevant studies. All articles that could meet the inclusion criteria as identified by either of the reviewers were selected.
Assessment of risk of bias (methodological quality)
Two reviewers (Elizabeth Martinez, Germana Gregorio), blinded to the origin of the papers, independently assessed the risk of bias (methodological quality) of identified studies using generation of allocation sequence, allocation concealment, blinding, and loss to follow up; and we recorded this information on a standard form.
We considered generation of allocation sequence to be: adequate if the study authors stated that they used a method resulting in unpredictable sequences, such as a random number table or list, or computer-generated random numbers; unclear if a trial was stated to be randomized but no further information was provided; or inadequate where allocation could be related to prognosis and therefore introduce selection bias, for example, date of birth or date of admission to hospital.
We considered allocation concealment to be: adequate if the assignment to arms of the study could not be predicted by the investigators or participants, for example, central randomization or numbered, identical drug containers; unclear if the study authors did not describe the method used to achieve concealment; or inadequate if they used a method such as alternation where the allocation of participants could be predicted.
We considered blinding to be: adequate when studies were double blind, that is, an identical placebo was used and recruitment to intervention or control arms was not known by either the investigator or the participants; unclear when the study authors provided no details; or open when they did not use blinding.
We considered loss to follow up to be: adequate when study endpoints were presented for 90% or more of the participants enrolled at the beginning; inadequate when follow up was less than this; or unclear when either or both the number of participants recruited at the beginning of the study and the number of participants who completed the study were not clear.
LD resolved disagreements regarding the assessment of methodological quality.
Data extraction
SA and Brown Okoko independently extracted data using standard forms. Key data items were aetiology and duration of diarrhoea, details of probiotic organism, participant characteristics (nutritional and HIV status), location (countries classified according to mortality stratum; WHO 2001), and the outcome measures listed above.
Data analysis
We pooled data from studies that used comparable outcome measures. For the duration of diarrhoea and number of stools per day of intervention, we achieved a pooled estimate of treatment effect by calculating the mean difference. For number of participants with diarrhoea lasting 3 days or more, or 4 days or more after starting the intervention, we calculated a pooled estimate of the risk ratio (RR) among probiotic and non-probiotic groups. We used either a fixed effect or random effects model approach according to the heterogeneity in outcomes across studies assessed by the chi-squared (chi
Where there was statistically significant heterogeneity in outcomes across studies, we conducted sensitivity analyses according to each of the four parameters of trial methodological quality.
We expected that differences between studies in probiotic(s) used, types of participants, and major causes of diarrhoea would result in heterogeneity in results. To address this likely heterogeneity, we conducted the following subgroup analyses.
- Probiotic type.
- Identified diarrhoeal pathogens (eg rotavirus).
- Background mortality rate (trials classified according to mortality stratum for children and adults in the country or countries where the trial was undertaken (WHO 2001) because of likely regional differences in major diarrhoeal pathogens related to the availability of clean water and level of sanitation).
- Age of participants.
Results
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification.
Our search identified 64 potentially relevant studies. Of these, 23 met the inclusion criteria and we excluded 35 (see 'Characteristics of excluded studies'). We have not been able to locate the reports of four potentially relevant studies to date and are awaiting a translation of one study (Taborska 1997) and the report of an unpublished study (Salazar-Lindo).
Publication status
Of the 23 included studies, 4 were published in the 1970s and 1980s, 18 in 1990 or later, and 1 is unpublished.
Study location
Two studies were performed in countries classified by the WHO as having high child and adult mortality, five in countries with low child and adult mortality, and one in a country with low child and high adult mortality (Shornikova 1997a), 14 classified as very low child and adult mortality, and 1 multicentre study pooled data from countries classified as having high and very low mortality rates (Guandalini 2000).
Sixteen studies were conducted in a single centre, five studies recruited participants from two or more centres within the same country, one study was a multicentre international study, and the number of recruitment centres was unclear in one study (D'Apuzzo 1982).
Participants
The 23 selected studies recruited a total of 1917 participants. There were 1449 infants or children (age < 18 years) divided into the probiotic (n = 740) and control (n = 709) groups, and 352 adults (age ≥ 18 years; 173 probiotic/179 control groups). Bruno 1983 studied participants aged 14 years and above, and participants in Wunderlich 1989 had a mean age of 33 years (age range not stated); these two studies accounted for 58 participants in each of the probiotic and control groups.
Fifteen studies recruited inpatients, three recruited outpatients, and three recruited both inpatients and outpatients; it was unclear in two studies whether the participants were inpatients and/or outpatients (Cetina-Sauri 1994; D'Apuzzo 1982).
Although all studies recruited participants with acute diarrhoea, the criteria for diarrhoea and for the duration of acute diarrhoea varied considerably between studies (see 'Characteristics of included studies'). Three or more loose or watery stools in the last 24 hours was the most common definition of diarrhoea (six studies); nine studies did not specify the definition of diarrhoea. The most commonly used criteria for the duration of acute diarrhoea was diarrhoea of < 5 days (four studies) or < 7 days (five studies), but criteria varied from < 48 hours to < 14 days, and the maximum duration of diarrhoea was not specified in eight studies.
No study specifically recruited or excluded travellers, and none identified any of the participants as suffering from travellers' diarrhoea. No study specifically recruited participants known to have HIV infection; no study stated HIV positivity as an exclusion criterion, but many excluded participants with chronic illness and/or immunosuppression.
Two studies specifically recruited malnourished children (Bhatnagar 1998; Raza 1995), and a further two studies included malnourished children (Pant 1996; Simakachorn 2000). Three studies excluded severely malnourished children (Carague-Orendain; Oandasan 1999; Raza 1995), and one excluded moderately malnourished children (Guarino 1997). Many studies did not comment specifically on nutritional status but excluded participants with underlying severe or chronic illness.
The hydration status of participants varied considerably between studies. Three studies included infants and children with severe dehydration (> 10%; Guandalini 2000; Pant 1996; Raza 1995), another 3 studies were restricted to infants and children with moderate (5 to 10%) or mild (< 5%) dehydration (Boulloche 1994; Guarino 1997; Simakachorn 2000), and 4 studies stated the average degree of dehydration (Isolauri 1994; Shornikova 1997a; Shornikova 1997b; Shornikova 1997c). Three studies excluded participants with moderate or severe dehydration (Buydens 1996; Cetina-Sauri 1994; Oandasan 1999). In the 4 studies that classified infants and children according to hydration status, 82 with no dehydration, 205 with mild, 123 with moderate, and 29 with severe dehydration were recruited (Guandalini 2000; Pant 1996; Raza 1995; Simakachorn 2000).
Ten studies specified that they excluded participants with bloody stools or grossly bloody stools, including three studies that were restricted to participants with rotavirus diarrhoea (Isolauri 1994; Shornikova 1997b; Sugita 1994). Participants with bloody stools were included in four studies. Nine studies did not state whether they excluded participants with dysentery. Four studies reported outcomes for the subgroup of children with rotavirus diarrhoea (Boulloche 1994; Guandalini 2000; Guarino 1997; Shornikova 1997a). Guandalini 2000 reported outcomes for the subgroup of children with bacterial diarrhoea, and Pant 1996 reported results for the subgroup of children with watery, non-bloody diarrhoea.
Interventions
Several different probiotics were tested: all were lactic acid bacilli, except in two studies that tested the yeast Saccharomyces boulardii (Cetina-Sauri 1994; Hochter 1990). Two studies tested a heat-killed probiotic preparation (Boulloche 1994; Simakachorn 2000). Few studies reported specific identification details beyond the species name, for example, by stating a culture collection number, and few undertook analyses to confirm the identity or viability of the organism. There was also wide variation in the treatment regimens according to number of organisms administered, timing of intervention, means of administration, and duration of treatment. Probiotics were administered directly to the participants or mixed with a variety of fluids and foods. Although expressed breast milk was used to administer probiotics in some studies, two studies excluded breastfed infants to minimize interruption of normal feeding.
Two studies compared the probiotic to standard treatment without a placebo (Guarino 1997; Isolauri 1994). One study administered milk formula to the comparison group (Bhatnagar 1998). The remaining 20 studies used a matching placebo.
Outcomes
In addition to the outcome measures identified a priori, we also extracted data for the number of participants with diarrhoea lasting 3 days or more and 4 days or more following the intervention. These outcomes were often reported and allowed several studies that did not report data for the other predefined outcomes to be included.
There was great variation between studies in the definition for the resolution of diarrhoea. The most commonly used criteria were the appearance of the last liquid or watery stool (five studies) or the first formed stool (four studies). Several studies used various combinations of stool consistency and stool frequency, and three studies included resolution of associated symptoms (fever, abdominal pain) in the definition. The criteria for resolution of diarrhoea were not stated in three studies. Raza 1995 did not follow children until resolution of diarrhoea but reported stool frequency by day of intervention.
There was also considerable variation between studies in outcome measures, and many studies reported more than one outcome. The most commonly reported outcomes were the number of participants with diarrhoea lasting 3 or more days (15 studies) and 4 or more days (13 studies) following the intervention, and the mean duration of diarrhoea from the start of the intervention (12 studies). Less commonly reported outcomes were weight change at intervals following the intervention (as a measure of rehydration), stool frequency according to day of intervention, length of hospital stay, and stool output. Although there was great variability between studies in definitions and reported outcomes, individual studies used the same criteria and outcomes for the probiotic and control groups.
Risk of bias in included studies
Methodological quality varied considerably (Appendix 1). Generation of the allocation sequence was adequate in 10 studies, inadequate in 1 study, and unclear in 12 studies. Concealment of allocation was adequate in only 5 studies, inadequate in 1 study, and unclear in 17 studies. Blinding was adequate in 14 studies, unclear in 6 studies, and 3 studies did not use blinding. Loss to follow up was adequate in 14 studies, inadequate in 7 studies, and unclear in 2 studies. Three studies were adequate for all of the four methodological quality assessment parameters (Oandasan 1999; Shornikova 1997c; Simakachorn 2000). Only one study stated that they conducted an analysis by "intention to treat" (Buydens 1996).
Effects of interventions
The results of trials according to reported outcomes are shown in Analysis 1.1 to Analysis 1.5 with trials grouped according to the probiotic(s) tested. With the exception of a trial of live Streptococcus thermophilus and Lactobacillus bulgaricus (Bhatnagar 1998; Analysis 1.4 and Analysis 1.5), all trials reported for all outcomes a beneficial effect in the probiotic group compared to the controls, and this was statistically significant in many studies. For several studies reporting continuous outcomes, standard deviations were large in comparison to mean values, suggesting that outcome data were not normally distributed. Therefore, caution is needed in the interpretation of the pooled results.
Diarrhoea lasting 3+ and 4+ days
3+ days
The persistence of diarrhoea on day 3 of intervention was the most frequently reported outcome (1341 participants/15 studies; Analysis 1.1). Participants taking probiotics were less likely to have diarrhoea lasting 3+ days, but there was heterogeneity between studies (0.67, 95% CI 0.61 to 0.74 (fixed effect model); 0.66; 95% CI 0.55 to 0.77 (random effects model; chi
4+ days
The risk ratio of diarrhoea lasting 4 or more days in the probiotic group was 0.36 (95% CI 0.30 to 0.44), again with heterogeneity in results between studies ( Analysis 1.2). The reduced risk of diarrhoea in the probiotic group was similar in a random effects analysis but had a wider confidence interval (0.31; 95% CI 0.19 to 0.50). Again, results were more homogeneous when the analysis was restricted to 4 studies with adequate allocation concealment ( Analysis 3.2), and this analysis showed a greater reduction in diarrhoea (RR 0.12; 95% CI 0.07 to 0.22). Heterogeneity between studies persisted in sensitivity analyses for the other parameters of study methodological quality. Differences in allocation concealment may have accounted for the heterogeneity in results between studies for the risk ratio of ongoing diarrhoea on days 3 and 4 of intervention. However, similar or greater effects of probiotics on stopping diarrhoea were seen when the analysis was restricted to studies with adequate allocation concealment.
Duration of diarrhoea
Mean duration of diarrhoea was reported in 12 studies, all performed in infants and children, and was reduced by 29.20 hours in people taking probiotics (95% CI 25.14 to 33.25, fixed effect model; 30.48 hours, 95% CI 18.51 to 42.46, random effects model, chi
Stool frequency
Participants in the probiotic group had on average 1.51 fewer stools on day 2 of intervention (95% CI 1.17 to 1.85; Analysis 1.4) and 1.31 fewer stools on day 3 (95% CI 1.07 to 1.56; Analysis 1.5) compared to participants in the control group. Although the results across studies were not statistically heterogeneous, mean stool frequency on day 2 was reported in only 5 trials and on day 3 in only 4 trials.
Need for unscheduled intravenous rehydration after randomization
Occasionally children developed severe dehydration and required intravenous rehydration (see 'Adverse events' below), but in no case was this attributable to probiotic treatment.
Death
No trial reported any deaths amongst participants.
Exploration of heterogeneity
We have explored methodological quality of studies as a potential source of heterogeneity in the primary analyses above and explored other pre-specified factors below.
Probiotic type
There were three or more trials of L. casei strain GG and Enterococcus LAB strain SF68 that reported the same outcomes. Two of three pooled analyses showed statistically significant between-study heterogeneity ( Analysis 1.2 and Analysis 1.5). Although the combination of live Lactobacillus acidophilus and Lactobacillus bifidus appeared to be particularly effective in reducing diarrhoea ( Analysis 1.2 and Analysis 1.4), this combination was tested in few participants. In contrast to most other regimens, live S. thermophilus and L. bulgaricus appeared to have no effect on diarrhoea, but this combination was tested in one trial only and the confidence width is wide and does not exclude potentially clinically significant values (Bhatnagar 1998; Analysis 1.1 and Analysis 1.2).
Identified diarrhoeal pathogen
Mean duration of diarrhoea in the subset of children with rotavirus diarrhoea was reported in two trials (Guandalini 2000; Guarino 1997) and in two studies that recruited only children with rotavirus diarrhoea (Isolauri 1994; Sugita 1994). Duration of diarrhoea was reduced by 38.10 hours (95% CI 8.10 to 68.10) in the probiotic group compared to the controls (random effects model; Analysis 5.1) − similar to that observed in the analysis of all-cause diarrhoea ( Analysis 1.3). Again there was marked between-study heterogeneity in results. Isolauri 1994 also reported that in children with rotavirus diarrhoea, the risk ratio of diarrhoea lasting 3 or more days was markedly reduced (0.22; 95% CI 0.05 to 0.91; Analysis 1.1). However, this analysis was based on relatively few participants. Guandalini 2000 reported that in rotavirus diarrhoea, stool frequency on day 3 of intervention was lower in children receiving L. casei strain GG (0.4, n = 56) than in the controls (2.0, n = 45; P < 0.05), which would appear to a greater reduction than seen in all-cause diarrhoea. Shornikova 1997a reported statistically significantly (P = 0.02) fewer watery stools in 13 children receiving L. casei strain GG compared with 21 receiving placebo, but presented no data. Boulloche 1994 commented that killed L. acidophilus reduced the duration of rotavirus diarrhoea to a similar extent as diarrhoea due to other causes.
Only two trials reported outcomes for participants confirmed to have bacterial diarrhoea, and both tested L. casei strain GG (Guandalini 2000; Shornikova 1997a). Both trials reported that diarrhoea was not reduced in the probiotic group compared to the control group.
Background mortality rate
We used the WHO mortality rates to reflect likely differences in the major causes of diarrhoea as a consequence of differences in the availability of clean water and level of sanitation in countries where trials were undertaken. We excluded the multicentre study by Guandalini 2000 because participants were recruited from countries with very low child and adult mortality and Egypt, which has high child and adult mortality. Only three trials were undertaken in countries with either high child or adult mortality (Bhatnagar 1998; Raza 1995; Shornikova 1997a). We observed statistically significant between-study heterogeneity in results where there were four or more trials in each classification ( Analysis 6.1 to Analysis 6.5). There was no consistent trend in the efficacy of probiotics according to this classification.
Age of participants
Statistically significant between-study heterogeneity was present for both the trials of adults and those of infants and children for the risk of diarrhoea lasting 3 or 4 or more days ( Analysis 7.1 and Analysis 7.2). We observed similar estimates of the effect of probiotics using the random effects model. Probiotics tended to reduce diarrhoea more in adults than children ( Analysis 7.1 to Analysis 7.4), especially for the risk of diarrhoea lasting 4 or more days ( Analysis 7.2). All trials reporting the mean duration of diarrhoea were conducted in infants and children.
Statistically significant between-study heterogeneity persisted in these subgroup analyses. Variability between studies may have persisted within subgroups. For example, the number of organisms and means of administration differed markedly even in studies of the same probiotic (see 'Characteristics of included studies'), and major causes of diarrhoea may have differed between countries with similar mortality rates. Also, other factors may have contributed to the between-study heterogeneity in the main analyses. Appendix 2 lists differences between studies in factors that may have influenced the progression of diarrhoea, such as antibiotic treatment before recruitment and nutritional differences in participants.
Adherence
Adherence to interventions was reported in only a few studies. Bhatnagar 1998 reported that in malnourished children the amount of milk consumed (control group) was statistically significantly greater than the amount of yoghurt (probiotic group). Boulloche 1994 reported that five participants receiving the probiotic and seven receiving the placebo did not comply with the study medication. Guandalini 2000 reported that one child in the probiotic group and three in the placebo group refused oral solutions. Rosenfeldt 2002a reported that four participants in the control group and four in the probiotic group were non-compliant with the trial protocol, and Rosenfeldt 2002b reported that 1 participant in the control group was non-compliant with the trial protocol, but no further details were given in either study.
Adverse events
Of all 23 selected studies, 12 studies reported that clinical observation of the participants revealed no adverse events, 8 did not collect or report information on adverse events, and 3 studies reported that an adverse event occurred (Pant 1996; Raza 1995; Shornikova 1997c).
Vomiting was reported as an adverse event; adverse events that led to withdrawal from treatment are described under 'Withdrawal from trial'. Pant 1996 reported that 1/19 children in the control group vomited 1 dose of the medication, but no vomiting occurred in the 20 children in the probiotic group. Raza 1995 reported that the frequency of vomiting on the second day of intervention was statistically significantly less in children in the probiotic than the placebo group. Shornikova 1997c reported that fewer children in the probiotic than the placebo group had vomiting from the second day of treatment and this was statistically significant on days 2 and 4. No child in the probiotic group vomited after the third day of treatment whereas vomiting persisted to the sixth day in 2/21 children in the placebo group.
No authors reported an adverse effect that they considered to be attributable to the probiotic.
Withdrawal from trial
Cetina-Sauri 1994 excluded from the analysis participants who deteriorated, developed concomitant illness and needed other drugs, or who wished to withdraw from the study. However, the number of participants withdrawn was not stated. In one multicentre study, it was not clear whether withdrawals occurred at the participating research centres (Guandalini 2000).
The study of L. casei strain GG in malnourished children reported that four children were removed from the study (Raza 1995); two siblings with cholera developed severe dehydration (one in each intervention group), and one of a pair of twins developed pneumonia and the other refused anything by mouth (one in each intervention group). Also, myoclonic jerks occurred in one participant receiving L. casei strain GG and one receiving placebo; both had severe dehydration on admission. Rosenfeldt 2002a excluded participants after randomization because antibiotics were prescribed (3 control group/2 probiotic group), rapid recovery occurred before intervention started (3 control group/1 probiotic group), or non-compliance with the trial protocol (4 control group/4 probiotic group). Rosenfeldt 2002b excluded participants after enrolment because of hospitalization with excessive vomiting and moderate dehydration (2 placebo group/3 probiotic group), because antibiotics were prescribed (1 placebo group) and because of non-compliance with protocol (1 placebo group). Shornikova 1997c withdrew one child from the placebo group after stool culture identified the probiotic under trial. Wunderlich 1989 withdrew three participants from the probiotic group and three from the control group on or after day 4 of the intervention, but stated that this was for reasons unrelated to medication. No other trial reported that participants in either probiotic or placebo groups had been withdrawn from the study. Therefore, where reported, in total 15 participants were withdrawn from the probiotic and 20 from the control groups.
Discussion
The striking finding of this review is that nearly all trials reported that probiotics had a beneficial effect in reducing diarrhoea, and this was statistically significant in many studies. This is despite great variability between studies in setting, participants recruited, probiotic tested, treatment regimens, and definitions of outcome measures. Given this marked variability in study design, it is not surprising that results varied markedly between studies. Although there was great variability in the methodological quality of the trials, there was no evidence that poor study design had led to an overestimate of the effects of probiotics. The small number of studies limited the ability to assess whether other factors may have accounted for between study heterogeneity, especially with regards to the probiotic(s) used and identified diarrhoeal pathogens. However, it did not appear that differences between studies in either regional differences in major pathogens or the age of participants were responsible for the heterogeneity in results. Between-study heterogeneity may have been due to the many other differences between studies, such as differences between participants according to prior antibiotic treatment, nutritional status and proportion with bacterial diarrhoea, and marked differences in probiotic dosages and methods of administration (see 'Characteristics of included studies' and Appendix 2).
This statistically significant between-study heterogeneity for nearly all reported outcomes indicates that the summary analyses of treatment effect should be interpreted with caution. However, when we conducted the analysis using the random effects model, which accounts for between study differences, we found broadly similar estimates of treatment effect.
There was a scarcity of information regarding the effects of probiotics for specific infectious agents. L. casei strain GG may be particularly effective for rotavirus diarrhoea, but more data are needed. Interpretation of the results of the subgroup analyses, in which studies were classified according to national mortality rates, is complicated because it is not clear whether this classification reflects important regional differences in the major causes of diarrhoea. However, it appeared that the effects of probiotics were similar in developing and developed countries. Probiotics may be more effective in acute diarrhoea in adults than infants and children.
The adverse event of vomiting was reported in three studies, all of which recruited children (including some malnourished children). Because vomiting is common in children with acute diarrhoea and it occurred less frequently in the probiotic than the control groups, it seems unlikely to be caused by the probiotics. The causes of the withdrawal of participants from trials appeared to be related mostly to their primary illness rather than the interventions. The reasons for non-compliance with protocol in some studies were not stated, but were unlikely to be related to the adverse events of probiotics since similar numbers of participants in the probiotic and control groups failed to comply. No authors reported adverse events that they considered to be attributable to probiotics. However, with the exception of malnourished children, most studies recruited previously healthy people. Therefore, no conclusions can be drawn regarding the safety of probiotics in other groups, for example, immunocompromised individuals, from this review.
Overall, we suggest that a variety of probiotics reduced infectious diarrhoea in children and adults in various settings. This suggests that a mechanism common to most probiotics, for example, colonization resistance, is effective against a wide range of gut pathogens. There were insufficient studies of specific probiotic regimens in defined groups of children or adults to allow for the development of definitive treatment guidelines. More well-designed studies are needed to advance the understanding of the efficacy of individual probiotics. Although trials of different probiotics in different participant groups and settings are to be encouraged, standardization of definitions of acute diarrhoea, treatment regimens, inclusion criteria, and outcome measures are needed to compare results across studies. All studies should try to present data separately for important subgroups, for example, according to identified causes of diarrhoea such as rotavirus or bacterial causes, or whether participants had received an antibiotic before recruitment. Guidance on undertaking trials with probiotics, such as reliably identifying the agent used, testing the viability of organisms and confirming their quantity, is readily available (Reid 1999).
Since most episodes of acute diarrhoea are uncomplicated, self limiting, and require no specific treatment, cost-effect analyses need to determine whether probiotics should be used in particular patient groups. For example, the apparent efficacy of probiotics in reducing the duration of acute diarrhoea may be particularly important in developing countries where acute diarrhoea in children is a risk factor for persistent diarrhoea (> 14 days) which, in turn, is closely associated with malnutrition (Walker-Smith 1993).
Authors' conclusions
Implications for practice In individual studies, probiotics appeared to be moderately effective as adjunctive therapy in reducing the duration of diarrhoea. However, there were insufficient studies of specific probiotic regimens in defined groups of children or adults to inform the development of evidence-based treatment guidelines. Most studies were performed in healthy individuals living in industrialized countries with uncomplicated diarrhoea, but some studies included malnourished children living in developing countries. |
Implications for research More studies of specific probiotic regimens in well-defined patient groups are needed to inform their role in clinical management. Trials need to use standardized definitions for acute diarrhoea and resolution of the illness, and they need to present data separately for important participant subgroups. All studies should include reliable identification of the probiotic being tested and confirm viability and number of organisms for live probiotics. Researchers should report whether the probiotic prevented or reversed dehydration − the most important complication of acute diarrhoea. In particular, the safety and potential of specific regimens to reduce the risk of persistent diarrhoea and associated malnutrition in children with acute diarrhoea in developing countries merits further study. Basic research is needed to determine the mechanisms underlying the apparent beneficial effects of probiotics in acute diarrhoea. |
Acknowledgements
Andreas Hansmann, Caroline Pedneault, Ludo Jobe, Saihou Sabally, Ai Koyanagi, and Christianna Morgan for translation of foreign language manuscripts. Hania Szajewska for providing manuscripts and translations. Paul Garner for overall guidance and advice.
Data and analyses
- Top of page
- Background
- Objectives
- Methods
- Results
- Discussion
- Authors' conclusions
- Acknowledgements
- Data and analyses
- Appendices
- What's new
- History
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Appendices
Appendix 1. Risk of bias (methodological quality) assessment
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Appendix 2. Summary of study design factors relevant to pooling data
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
What's new
Last assessed as up-to-date: 19 June 2003.
| |||||||||
History
Protocol first published: Issue 2, 2001
Review first published: Issue 2, 2004
Contributions of authors
Stephen Allen, Leonila Dans, and Germana Gregorio identified articles for inclusion in the review. Leonila Dans and Elizabeth Martinez assessed study quality and Germana Gregorio settled any disagreements. Stephen Allen and Okoko Brown extracted data. Stephen Allen took the main responsibility for analysis and writing the review. All reviewers contributed to the final version.
Declarations of interest
Stephen Allen is participating in research with Lactobacillus casei strain GG, provided by Valio Ltd, Finland. Scientific Hospital Supplies UK provided this probiotic for previous studies of acute diarrhoea and also support to attend a training workshop.
Sources of support
Internal sources
- Medical Research Council Laboratories, Gambia.
- University of Oxford, UK.
External sources
- Department for International Development, UK.
Differences between protocol and review
2004, Issue 2 (first review version):
- Primary outcomes: "Diarrhoea lasting 3+ and 4+ days" added. Following review of the selected studies, it was clear that the proportion of participants with diarrhoea lasting 3 and 4 or more days after intervention was reported in many studies. Therefore, these outcome measures were included in the meta-analysis.
- Secondary outcomes: "side effects" and "vomiting" replaced by "adverse events". Data were extracted on adverse outcomes that occurred during trials whether or not their causality was assessed. "Need for unscheduled intravenous rehydration after randomization" and "Deaths" are now secondary outcomes.
Index terms
Medical Subject Headings (MeSH)
Diarrhea [microbiology; parasitology; *therapy]; Probiotics [*therapeutic use]
MeSH check words
Adult; Child; Humans
* Indicates the major publication for the study