Acute diarrhea is an important cause of morbidity and mortality among children in developing countries throughout the world, particularly during the second year of life (1). According to World Health Organization (WHO) recommendations, clinical management of acute diarrhea is largely symptomatic and involves 2 main steps: replacement of fluid and electrolyte losses and nutritional therapy with early repeat feeding (2).
The oral rehydration salts (ORS) advocated by the WHO have been widely used in developing countries and are now well known at all social levels. As a result, dehydration-related mortality has decreased considerably in young children during the past few decades (3). However, ORS do not reduce the volume or duration of watery diarrhea (4) and their use is hampered because they are not considered to be a medical treatment. Therefore, a significant proportion of diarrhea episodes involve the use of unnecessary antimicrobial drugs (5,6). It is therefore important to find effective and safe adjuvant treatments to decrease the length and severity of the diarrhea.
Rotavirus is the most common pathogen of infectious watery diarrhea. It is responsible for 29% to 45% of cases of severe diarrheic episodes in developing and industrialized countries (7), followed by enterotoxigenic Escherichia coli and other bacteria (8). Clinical reports have documented the therapeutic effectiveness of lactic acid bacteria when used as a probiotic in the treatment of acute diarrhea of viral origin (9,10). The activity of Lactobacillus species against pathogenic organisms in the intestine is believed to occur through any of the following mechanisms: competition for active sites on the enterocyte, cellular protection such as the creation of biofilms to prevent access of enteric microorganisms, and production of antimicrobial substances (11).
In the present study we have chosen to examine the therapeutic effectiveness of an antidiarrheal product containing a heat-killed Lactobacillus LB strain (Lacteol Sachet; Axcan Pharma, Houdan, France) as an adjunct to ORS. Although Lactobacillus LB has been widely used, there are no substantial medical reports regarding its in vivo effectiveness. Experimental evidence of the antimicrobial activity of this biotherapeutic agent against diarrhea-causing organisms has been provided in the past 10 years in cellular and animal models (12–17). The use of lyophilized, heat-killed Lactobacillus LB with its spent culture supernatant in the treatment of acute diarrhea has recently been reported effective in promoting a more rapid recovery from acute, watery diarrhea in children, with reduced morbidity, reduced time spent in the hospital (when hospitalization is necessary), and reduced parental care required in managing acutely sick children (18).
In this article we report the results of a double-blind, placebo-controlled, randomized multicenter study carried out in a population of children with acute watery diarrhea with the purpose of confirming the effectiveness of Lactobacillus LB compared with a placebo formulation. The objective of this study was to determine whether Lactobacillus LB, in association with oral rehydration, would decrease the duration of diarrhea compared with placebo. The study's secondary objectives were to evaluate the safety as well as the medical and economic advantages of treatment with Lactobacillus LB.
PATIENTS AND METHODS
This was a phase III multicenter, randomized, double-blind study with 2 parallel groups (Lactobacillus LB and placebo). It was conducted over a period of 4 months (September 2004–December 2004) in 4 participating centers in Lima, Peru. The same documentation was used in each center. Pretrial training of study personnel in each center was carried out and the study was closely monitored by a trained local monitor. The data were collected by investigators using a case report form. The forms were sent directly to a third party for data management, statistical analysis, and analysis reporting. Study codes were broken only after completion of the blind review. The study was carried out with 80 children between 3 months and 4 years of age in 2 balanced groups of 40 children each. The study protocol and consent form used in this study were reviewed and approved by the institutional ethical committee of each participating center. Each patient was treated until recovery or for a maximum of 4.5 days.
After screening, the parents or legal guardians of eligible patients were invited to participate in the study. Children whose parents or legal guardians provided signed consent were randomized to receive an initial dose of 2 sachets of placebo or Lactobacillus LB and to continue treatment at home, with 1 sachet every 12 hours for 4.5 days (10 sachets on completed treatment). Parents were provided with a diary to record detailed information on stool characteristics and frequency. A medical visit was scheduled for day 1 and at the end of the 4.5 days (ie, 108 hours) of the study, when all unused medication and the report diary were collected. Another complete physical examination was performed at that time.
Children were selected for the study if they presented at the hospital with a history of acute diarrhea of presumably infectious origin, with a duration of <72 hours and with 3 or more watery stools within the previous 24 hours. Children were ineligible for the study if they presented with signs of dehydration requiring hospitalization according to WHO guidelines (2), bloody stools, chronic gastrointestinal disease (eg, cystic fibrosis or celiac disease), a chronic immunological condition that could potentially cause diarrhea (eg, AIDS), or a lactose or fructose intolerance. Other exclusion criteria included hemodynamic abnormalities, neurological disturbance, rectal body temperature >39.0°C, and previous treatment with antibiotics or a drug interfering with intestinal motility. On admission to the study the child's medical history and personal information were obtained, as well as a full physical examination including weight, height, and rectal temperature. A stool sample was obtained and tested for rotavirus with an enzyme-linked immunosorbent assay (Premier Rotaclone Meridian Bioscience, Cincinnati, OH).
Fluid losses from diarrhea were replaced with Viatol (Axcan Pharma), an oral rehydration solution containing 111 mmol/L glucose, 12 mmol/L citrate, 50 mmol/L chloride, 50 mmol/L sodium, and 24.5 mmol/L potassium. The active treatment administered was Lacteol 340 mg powder for oral suspension in single sachet doses. Each patient in the treatment group received 20 billion units of killed Lactobacillus LB strain and 320 mg of neutralized supernatant spent culture medium per day (in sachets containing 10 billion microbes each). The placebo was composed of salicylic acid, banana and orange flavor, sucrose, and yellow and brown iron oxides.
Lacteol and the placebo were manufactured, packaged, and provided by Axcan Pharma, Houdan, France. Viatol was manufactured, packaged, and provided by Sophartex Laboratories, Vernovillet, France.
The primary endpoint was the duration of diarrhea, measured as the time between inclusion and the end of the diarrhea episode. The end of the diarrhea episode was defined as the time to the first normal stool followed by 2 consecutive normal stools or the time to the last diarrheic stool followed by 12 hours without stool.
The analyses were performed on all of the patients and on both age subgroups: 3 months to 2 years of age and 2 to 4 years of age. Complementary statistical analyses were performed to determine the influence of age range, duration of the diarrhea episode before inclusion, and rotavirus status. Secondary criteria analyzed were the change in weight between hour 0 and the final weight assessment, and the time between hour 0 and the last diarrheic stool before recovery.
Data management and statistical analysis were carried out by UMANIS Clinical Research in France, an independent clinical research organization, in accordance with the analysis plan based on the protocol and finalized before study unblinding. Axcan Pharma had no role in the analysis of data from or during the study.
Sample size populations were determined based on previous studies and according to established parameters: α value of 0.05, β value of 0.2, and a minimum difference of 24 hours between the 2 groups. The minimum number of patients per group was 33.
Descriptive statistics were provided according to the nature of the variables for each treatment group and for the whole population. For the survival analyses, the data of the patients with unresolved diarrhea were considered as censured data (108 h). For 2-sided tests, the type I error risk of statistical tests was set at 5%. Quantitative parameters were analyzed by a Student t test or a Mann-Whitney Wilcoxon test depending on data normality. Normality was assessed by the Shapiro-Wilk test. Qualitative parameters were analyzed by a χ2 test if the frequency of each class was >5; otherwise, a Fisher exact test was used. Kaplan-Meier curves were displayed for both treatment groups for the analysis of duration of diarrhea. An adjusted Cox proportional-hazards model was used involving treatment effect and prognostic factors (age, duration of the episode of diarrhea at hour 0, and rotavirus status) to estimate the relative risk of placebo vs Lactobacillus LB treatment.
Eighty patients were randomized (40 were treated with Lactobacillus LB and 40 with placebo). All of the patients received at least 1 dose of Lactobacillus LB or placebo. There were 3 study dropouts after randomization, all in the Lactobacillus LB group: 1 due to an adverse event (severe dehydration) and 2 because of the parents' decision to withdraw their child from the study. There were no discontinuations in the placebo group. Of the remaining 77 patients, 6 patients (1 in the Lactobacillus LB group and 5 in the placebo group) continued treatment for 4.5 days and did not recover. Seventy-one patients recovered, 36 (50.7%) in the Lactobacillus LB group and 35 (49.3%) in the placebo group. The intent-to-treat patient set included 80 patients and was defined as all patients, in accordance with the inclusion and noninclusion criteria, having taken at least 1 dose of study medication with at least 1 assessment after inclusion. There were no study deviations from what was planned.
Overall Patient Profile and Baseline Characteristics
Children were between 3 and 42 months of age, with mean ages of 12.3 months in the placebo group and 16.5 months in the Lactobacillus LB group. The numbers of male patients were 18 in the control group and 26 in the Lactobacillus LB group. Mean height, weight, and body mass index were similar between groups.
The mean duration of diarrhea and mean number of watery stools in the 24 hours before inclusion was similar for the 2 groups. Only 18 of the 80 patients (22.5%) tested positive for presence of rotavirus, 12 in the treatment group and 6 in the placebo group. Rectal temperature ranged between 36.0°C and 38.5°C at inclusion. The descriptive statistics and clinical characteristics did not show any nonequivalence between groups for the set of criteria studied (Table 1).
The 2 groups were comparable in terms of previous treatment, with 65 patients (81.3%) in both groups having received no previous treatment. Five patients (6.3%) in the Lactobacillus LB group had received at least 1 treatment without antidiarrheal action.
Measurement of Treatment Compliance
The number of sachets taken was 2 at hour 0 and 1 every 12 hours thereafter. Compliance was determined by dividing the number of sachets used by the theoretical number of sachets prescribed by the study for the duration of treatment. For all of the patients the mean treatment duration was 34.7 hours (SD, 31.2; range, 0–128 hours), corresponding to 4.8 sachets. The mean sachet intake was of 5.2 sachets, for a mean compliance rate of 108.2% (SD, 14.2%; extreme values, 88.9%–180%). Compliance was >100% because parents could renew treatment if the child vomited in the hour after administration. The mean compliance in the subgroup with duration of diarrhea of more than 24 hours was 107.8% (SD, 12.3%).
Effectiveness Results: Duration of Diarrhea
Effectiveness was analyzed for all patients in each treatment group and for both designated age groups (3 months to 2 years and 2–4 years). For statistical analysis, the data of the patients with unresolved diarrhea were estimated using the total duration of the study (108 hours) and the duration of diarrhea in patients who withdrew from the study was considered to be missing.
In all of the patients and for both age groups, the median duration of diarrhea was shorter for the Lactobacillus LB group, but there were no statistically significant differences. From a clinical point of view, the reduction in duration of diarrhea was 6.6 hours (median duration of diarrhea of 16.6 h in placebo group vs 10.0 h in Lactobacillus LB group; P = 0.275). In patients 3 months to 2 years of age the clinically observed reduction in duration of diarrhea was 2.0 hours; in patients 2 to 4 years of age, the reduction was 20.5 hours (Table 2).
The rates of recovery (ie, end of the diarrhea episode) were compared between treatments in a survival analyses, and the data of the patients with unresolved diarrhea were considered to be censured data (108 hours). The patients withdrawn from the study were considered to be censured at their last assessment or at their last reported stool. The analysis was therefore performed on the 80 patients as a whole and separated by age groups. During the total duration of the study (108 hours), 71 patients had their diarrhea resolved, 36 in the Lactobacillus LB group and 35 in the placebo group. Survival curves did not show a statistically significant difference between the groups (P = 0.334). Fifty-seven of the 65 patients 3 months to 2 years of age were cured after 108 hours: 27 in the Lactobacillus LB group and 30 in the placebo group. There was no statistically significant difference between groups (P = 0.522). Fifteen patients were between 2 and 4 years of age; of these, 9 in the Lactobacillus LB group and 5 in the placebo group were cured. The survival curves were compared using the log-rank test, which did not show a significant difference between the groups (P = 0.468). An analysis of variance was performed on the intent-to-treat population by age (3 months to 2 years and 2–4 years) and rotavirus status, which did not show a significant statistical difference between the treatment groups.
Further analyses were performed by grouping the patients according to the duration of the episode of diarrhea before randomization (Fig. 1). In the subgroup of patients with a history of diarrhea of <24 hours before randomization, the median duration of diarrhea was shorter for the placebo group, without a significant difference. In the subgroup with a history of diarrhea of more than 24 hours (n = 58) there was a marked and statistically significant difference (P = 0.044) between treatment groups, with the median duration of diarrhea considerably shorter in the group treated with Lactobacillus LB (8.2 hours) than in patients treated with placebo (30.4 hours; Table 2). From a clinical point of view, the gain in terms of reduced duration of diarrhea was 22.2 hours in favor of Lactobacillus LB.
Secondary Outcome Criteria
There was no significant difference between treatment groups in terms of change in weight between the time of randomization and the final assessment (P = 0.946; median, 0.10 kg for Lactobacillus LB and placebo). There was no significant difference in the relative change in weight between groups (P = 0.763; median, 0.92 kg for Lactobacillus LB and 1.05 kg for placebo). The total intake of oral rehydration solution was similar for both groups throughout the study.
Safety was assessed mainly by considering vomiting and adverse events. Eighteen patients vomited at some point while receiving study medication: 12 (30%) in the Lactobacillus LB group and 6 (15%) in the placebo group. No significant difference was observed for vomiting between the 2 groups. Two patients, 1 receiving Lactobacillus LB and 1 receiving placebo, experienced an unexpected event while receiving study medication. The patient in the Lactobacillus LB group had severe dehydration and was withdrawn from the study. The patient in the control group developed an itchy rash that was probably a reaction to a compound in the placebo treatment such as flavoring.
Both treatment groups were statistically comparable with regard to their admission characteristics: age, sex, nutritional indicators, duration of diarrhea before inclusion, and illness severity as determined by the number of stools passed during the previous 24 hours. Treatment compliance in this study was remarkably high and the proportion of patients lost to follow-up was low (3 of 80 randomized patients).
The main study outcome was assessed with the participation of the patients' parents, who filled out a diary recording specific data on stools and diarrhea duration. From these diaries, study personnel had no difficulties in recognizing characteristics of stools passed by the study patients. Because no emergency occurred to necessitate breaking the study codes, the double-blind nature of this trial was maintained throughout the study.
All of the admission stools were examined for the presence of rotavirus particles with the use of enzyme-linked immunosorbent assay. The proportion of rotavirus-positive cases was only 22.5%, which differs from the average 40% prevalence level described in this kind of population in previous studies (19,20).
The natural course of mild diarrhea is spontaneous recovery in 3 to 5 days. The median duration of diarrhea after receiving the first dose of placebo or Lactobacillus LB was short (16.6 hours in the placebo group and 10.0 hours in the Lactobacillus LB group; P = 0.275); duration was not statistically different between the 2 treatment groups. However, when we compared the 2 treatment groups after excluding patients admitted with a duration of diarrhea no longer than 24 hours before admission, a significantly shorter duration of diarrhea was found in those treated with Lactobacillus LB. Median reduction of diarrhea duration attributed to Lactobacillus LB in this subgroup was 73.0% (30.4 h with placebo and 8.2 h with Lactobacillus LB; P = 0.044). In the placebo group recovery was observed after 3 days in the 24- to 48-hour and 48- to 72-hour subgroups, in accordance with the normal course of the disease. In the subgroup with diarrhea duration of 0 to 24 hours, recovery time was abnormally short for the placebo group, presumably because patients are simply experiencing an intestinal disorder rather than true diarrhea.
Because the study was conducted to include only patients who could be clinically managed on an outpatient basis, thus excluding those with more severe diarrhea necessitating in-hospital care, it is likely that a significant number of patients with very mild, short-lived diarrhea were included. Lactobacillus LB has shown high affinity in attaching to enterocytes, thus preventing colonization by Gram-positive and Gram-negative organisms. It also generates a heat-stable antimicrobial substance with anti-invasive and antiadhesive activity against intracellular pathogens (14–16). This may explain the significant effect of Lactobacillus LB on the clinical course of the illness in patients with diarrhea duration >24 hours before receiving treatment. The number of patients who did not recover was low (n = 1) in the Lactobacillus LB group compared with the placebo group (n = 5). In light of this result, it is possible that we reduced the need for hospital admission in certain patients.
In mild episodes of diarrhea (ie, a few hours of diarrhea without clinical signs of dehydration), it is likely that no medical intervention is necessary other than educating parents to provide more fluids and to recognize when the diarrhea is worsening or becoming overly prolonged. This study shows that patients with diarrhea sometimes recover spontaneously, which would also explain why Lactobacillus LB had no significant impact on the course of mild diarrhea.
The safety profile of Lactobacillus LB found in this study is good. No major side effects were noted, and although vomiting appeared to occur more frequently in the Lactobacillus LB group in this study than in others (9), the studies in the literature used different strains of Lactobacillus and study populations consisted of inpatients, presumably with more severe disease. More studies should be carried out, preferably on inpatient populations with well-established diarrhea, to better assess the impact of Lactobacillus LB on more severe cases.
In summary, the results of this study show that the mean duration of diarrhea in the intent-to-treat population is less for the treatment group compared with the placebo group. The advantage of Lactobacillus LB, although clinically important, is not statistically significant (P = 0.275).
In the 58 patients with duration of diarrhea of more than 24 hours at inclusion, there was a significant difference between the 2 groups in favor of Lactobacillus LB (P = 0.044). The effectiveness was clinically and statistically proven by a reduction in median duration of diarrhea of >22 hours. This result may indicate that Lactobacillus LB would be most effective in well-established diarrhea. Further studies in patients with more severe diarrhea should be performed to obtain a more complete assessment of the effectiveness of Lactobacillus LB.
The data on safety was excellent, with only 1 patient in each group experiencing an unexpected adverse event. With this new clinical trial, the effectiveness and safety of Lactobacillus LB is confirmed in the treatment of well-established (>24 h) mild acute diarrhea.
1. Kosek M, Bern C, Guerrant RL. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull WHO 2003; 81:197–204.
2. World Health Organization. The Treatment of Diarrhoea—A Manual for Physicians and Other Senior Health Workers (WHO/CAH/03.7). Geneva: World Health Organization 2004.
3. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year? Lancet 2003; 361:2226–2234.
4. El-Mougi M, El-Akkad N, Hendawi A, et al
. Is a low-osmolarity ORS solution more efficacious than standard WHO ORS solution? J Pediatr Gastroenterol Nutr 1994; 19:83–86.
5. Thapar N, Sanderson IR. Diarrhoea in children: an interface between developing and developed countries. Lancet 2004; 363:641–653.
6. Mittal SK, Mathew JL. Regulating the use of drugs in diarrhea. J Pediatr Gastroenterol Nutr 2001; 33(Suppl 2):S26–S30.
7. Parashar UD, Gibson CJ, Bresse JS, et al
. Rotavirus and severe childhood diarrhea. Emerg Infect Dis 2006; 12:304–306.
8. Nataro JP, Kaper JB. Diarrheagenic Escherichia coli
. Clin Microbiol Rev 1998; 11:142–201.
9. Van Niel CW, Feudtner C, Garrison MM, et al
. Lactobacillus therapy for acute infectious diarrhea in children: a meta-analysis. Pediatrics 2002; 109:678–684.
10. Allen SJ, Okoko B, Martinez E, et al
. Probiotics for treating infectious diarrhea. Cochrane Database Syst Rev 2004; 2:CD003048.
11. Isolauri E, Kirjavainen PV, Salminen S. Probiotics: a role in the treatment of intestinal infection and inflammation? Gut 2002; 50(Suppl III):III54–III59.
12. Chauviere G, Coconnier MH, Kerneis S, et al
. Adhesion of human Lactobacillus acidophilus
strain LB to human enterocyte-like Caco-2 cells. J Gen Microbiol 1992; 138:1689–1696.
13. Chauviere G, Coconnier MH, Kerneis S, et al
. Competitive exclusion of diarrheagenic Escherichia coli
(ETEC) from human enterocyte-like Caco-2 cells by heat-killed Lactobacillus. FEMS Microbiol Lett 1992; 70:213–217.
14. Coconnier MH, Bernet MF, Kerneis S, et al
. Inhibition of adhesion of enteroinvasive pathogens to human intestinal Caco-2 cells by Lactobacillus acidophilus
strain LB decreases bacterial invasion. FEMS Microbiol Lett 1993; 110:299–306.
15. Coconnier MH, Lievin V, Bennet-Camard MF, et al
. Antibacterial effect of the adhering human Lactobacillus acidophilus
strain LB. Antimicrob Agents Chemother 1997; 41:1046–1052.
16. Coconnier MH, Lievin V, Lorrot M, et al
. Antagonistic activity of Lactobacillus acidophilus
LB against intracellular Salmonella enterica
infecting human enterocyte-like Caco-2/TC-7 cells. Appl Environ Microbiol 2000; 66:1152–1157.
17. Lievin-Le Moal V, Amsellen R, Servin AL, et al
. Lactobacillus acidophilus
(strain LB) from the resident adult human gastrointestinal microflora exerts activity against brush border damage promoted by a diarrheoagenic Escherichia coli
in human enterocyte-like cells. Gut 2002; 50:803–811.
18. Simakachorn N, Pichaipat V, Rithipompaisam P, et al
. Clinical evaluation of the addition of lyophilized, heat-killed lactobacillus acidophilus LB to oral rehydration therapy in the treatment of acute diarrhea
in children. J Pediatr Gastroenterol Nutr 2000; 30:68–72.
19. Salazar-Lindo E, Miranda-Langschwager P, Campos-Sanchez M, et al
. Lactobacillus casei
strain GG in the treatment of infants with acute watery diarrhea: a randomized, double-blind, placebo controlled clinical trial. BMC Pediatr 2004; 4:18.
20. Salazar-Lindo E, Santisteban-Ponce J, Chea-Woo E, et al
. Racecadotril in the treatment of acute watery diarrhea in children. N Engl J Med 2000; 343(7):463–467.