Indian J Pediatr (April 2012) 79(4):478482 DOI 10.

1007/s12098-011-0573-z

ORIGINAL ARTICLE

Efficacy and Safety of Saccharomyces boulardii in Acute Childhood Diarrhea: A Double Blind Randomised Controlled Trial
Musheer Riaz & Seema Alam & Abida Malik & S. Manazir Ali

Received: 25 September 2010 / Accepted: 27 September 2011 / Published online: 14 October 2011 # Dr. K C Chaudhuri Foundation 2011

Abstract Objective To see the efficacy and safety of 250 mg of Saccharomyces boulardii twice daily for 5 d in acute childhood diarrhea. Methods Children aged between 3 mo and 59 mo with acute onset diarrhea (of less than 48 h) admitted in DTTU (diarrhea treatment and training unit) were included and those with clinical evidence of severe malnutrition, systemic infection, encephalopathy and/or convulsion, electrolyte imbalance, invasive diarrhea or previous use of any probiotics were excluded from the study. Those included randomly were given either a placebo or Saccharomyces boullardi (SB) in identical packets mixed with puffed rice powder. Results Mean post intervention duration of diarrhea was significantly (95% CI=28.13 to 5.43) shorter in SB group (52.0824.57 h) as compared to placebo group (64.0430.43 h). The time of appearance of first semi formed stool in SB group (39.4823.09 h) was significantly (95% CI 25.4 to 3.87) shorter than the placebo group (54.1328.21 h). No statistically significant difference was found in rest of the parameters.
M. Riaz : S. M. Ali Department of Pediatrics, Jawaharlal Nehru Medical College, AMU, Aligarh, UP, India S. Alam (*) Department of Pediatric Hepatology, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, India e-mail: seema_alam@hotmail.com A. Malik Department of Microbiology, Jawaharlal Nehru Medical College, AMU, Aligarh, UP, India

Conclusions There is initial evidence available that SB may have a therapeutic role in the management of acute childhood diarrhea. Keywords Acute diarrhea . Probiotics . Saccharomyces boularrdii . Therapeutic role

Introduction More than one billion episodes of diarrhea occur every year among children under 5 y of age causing approximately 2.5 million deaths [1, 2]. Education and widespread use of oral rehydration therapy has reduced the number of diarrhea deaths [1] but the morbidity still remains high. Various meta-analyses [35] on use of probiotics in treatment of acute childhood diarrhea showing beneficial effect of Lactobacillus GG. Saccharomyces boulardii has been studied [610], but mostly in developed countries. The etiological profile of childhood diarrhea in developing countries is different and only few studies with questionable methodology have been done in the developing world [7, 9, 10]. Hence, the authors planned the present study to see the efficacy and safety of 250 mg of Saccharomyces boulardii twice daily for 5 d in children aged 3 mo59 mo with acute childhood diarrhea.

Material and Methods This double blinded randomized controlled trial was conducted on children aged between 3 mo and 59 mo with acute onset diarrhea (of less than 48 h) admitted in DTTU (diarrhea treatment and training unit) of the Department of Pediatrics, Jawaharlal Nehru Medical College, AMU, from

Indian J Pediatr (April 2012) 79(4):478482

479

May 2008 through September 2009, after obtaining an informed consent from the parents/guardians/care givers. Institutional ethical clearance was taken for the study to get it registered at the clinical trial registry (Regd. No. UTRNWHO TEMP UTRN 06101562521112008353496/ CTRI No. and Date- TRI/2008/091/000255, 05-01-2009). All children with clinical evidence of severe malnutrition, systemic infection, encephalopathy and/or convulsion, electrolyte imbalance, invasive diarrhea or previous use of any probiotics were excluded from the study. All children with weight for age below 60% or weight for length/height below 70% were considered as suffering from severe malnutrition. And those with the weight for age between 70% and 60% and weight for height between 70% and 80% were considered as moderately malnourished. After informed consent the children were randomly given either a placebo or Saccharomyces boullardi (SB) in identical packets mixed with puffed rice powder. Children receiving SB were given commercially available Saccharomyces boulardii (Econorm, Dr Reddys Laboratories) spores from the same pharmaceutical company throughout the period of study. Each child was advised to take the packet containing either placebo or SB (250 mg), as given to him, twice daily for 5 d or till the recovery whichever was earlier. These children were managed as per the guidelines of the IAP task force on diarrhea and thus, the given packets were apart from low osmolarity ORS, Zinc and feeding advised to them. A non departmental colleague not involved in study randomized (block randomisation) these identical packets of placebo or SB. Data collection including complete admission characteristics, anthropometric measurement was recorded separately
Table 1 Comparison of baseline characteristics of study groups Characteristics

on a performa for each patient. Frequency of stools, hydration status, amount of ORS consumed, duration of diarrhea and weight recording was done every 4 h. Patients were considered as treatment failure if dehydration persisted for more than 72 h, if consumption of ORS was more than 8 l and if the need for unscheduled intravenous fluids was more than 200 ml/kg throughout hospital stay. The data of all those who left the study or were treatment failures was collected till the time they remained in the study and was included in the analysis. The primary outcome variables were duration of post intervention diarrhea (defined time from enrolment to recovery). Recovery and discharge criterion was defined as passage of 3 consecutive semi formed stools or no stools for 12 h. The semi formed stool was observed and collected by the mother. The stool was observed immediately as well as time noted by the health personelle. At least 23 ml of stool specimen was collected in on Cary Blairs transport media and sent for the isolation of Vibrio cholerae. Another stool specimen was collected for detection of rotavirus antigen in a clear, dry vial free of detergent residue and sterilized. All stool samples for detection of rotavirus antigen were stored undiluted at 20C of temperature till the test was performed. The rotavirus detection was done using an ELISA based kit. Statistical Analysis Data was analyzed with the help of SPSS (version 11.5). Continuous data were compared by student t test and the MannWhitney U test. The categorical variables were

SB group(n=54) Mean ( SD) 20.6716.82 27 10 10 75.8513.13 94.115.78 83.3910.72 21.1812.73 6.644.82 22.7820.49 22 26 06

Placebo group(n=54) Mean ( SD) 25.2217.72 34 13 14 72.8711.22 92.665.59 82.3110.82 18.0210.32 8.306.52 18.0915.94 23 24 07

95% CI for difference (11.14-2.04)

Mean age in months Age group (n) 312 mo 1336 mo 3759 mo Mean weight for age (%) Mean height for age (%) Mean weight for height (%) Mean frequency of stools in 24 h prior to admission Mean frequency of vomiting in 24 h prior to admission Mean duration of diarrhea prior to admission Hydration status (n) No dehydration Some dehydration Severe dehydration

0.174 0.394

0.211 0.190 0.602 0.159 0.149 0.187 0.914

(1.72-7.67) (0.73-3.62) (3.02-5.19) (1.26-7.59) (3.94-0.61) (2.32-11.71)

480 Table 2 Comparison of outcome variables between the two study groups Outcome variables SB group Mean SD n=54 3532.042462.48 26.7219.21 16.0817.37 [n=36] 15.2811.71 [n=14] 39.4823.09 [n=43] 7.9210.70 [n=50] 402.68417.76 [n=54] 919.16423.11 [n=12] 52.0824.57 [n=54]

Indian J Pediatr (April 2012) 79(4):478482

Placebo group Mean SD n=54 4400.372535.62 30.6222.17 17.6317.26 [n=41] 15.9213.43 [n=25] 54.1328.21 [n=47] 10.8917.63 [n=47] 492.55541.08 [n=54] 1174.73980.68 [n=19] 64.0430.43 [n=54]

95% CI for difference

Total ORS consumed (ml) No. of stools 024 h 2548 h 4972 h Time of appearance of first semi formed stool (h) Time of first rehydration (h) Weight gain (g) Total IVF needed (ml) Duration of post intervention diarrhea (h)

0.074 0.330 0.696 0.883 0.008 0.315 0.336 0.402 0.031

(1821.9-85.28) (11.8-4.01) (9.43-6.32) (9.32-8.06) [25.4-(3.87)] (8.81-2.86) (274.3-94.5) (870.5-359.34) [22.4-(1.11)]

analyzed with the help of Chi-square test or Fischers exact test. The required sample size was calculated to be 50 for experimental subjects. This was calculated to detect a difference of 25% of post intervention duration of diarrhea, using a value obtained from an earlier work from this centre of 31.114.24 h, at significance level of 0.05 and power of 80% for a study design using t-test. Using SPSS (version 11.5), t-test was performed and chi square analyses (Fishers exact) were done to compare proportions between the two independent groups.

Results Of the total of 990 cases who reported with acute diarrhea, 550 were in the age group of 3 mo to 5 y. Of them 308 were severely malnourished, 98 were having systemic infections and 36 did not give consent for the study. A total of 108 children aged between 3 mo and 5 y were included in the study. Of these 108 cases, 42 (19 in SB group and 23 in Placebo group) were moderately malnourished. Out of these 108 children, nine patients left the study and nine

patients fulfilled the criteria of treatment failure.Stool samples of all patients were sent for culture and rotavirus isolation in stools could be done for the initial 93 cases due to logistic constraints. Out of these 93 stool samples, 14 tested positive for rotavirus antigen. Baseline admission characteristics were comparable between the two groups (Table 1). Six cases in SB group and 3 cases in placebo group left the study. There were 5 and 4 treatment failures in the SB and placebo group (OR=0.732, 95% CI 0.18- 2.9). The mean post intervention duration of diarrhea was significantly (95% CI=28.13 to 5.43) shorter in SB group (52.0824.57 h) as compared to placebo group (64.04 30.43 h). The time of appearance of first semi formed stool in SB group (39.4823.09 h) was significantly (95% CI 25.4 to 3.87) shorter than the placebo group (54.13 28.21 h). No statistically significant difference was found in rest of the parameters (Table 2). But when those cases who left the study or were treatment failures were excluded, the authors found that the mean total ORS intake became significantly (95% CI 2004.4 to 208.3) lower and mean time of first rehydration became significantly (95% CI

Table 3 Comparison of outcome variables between the two study groups after removal of cases who left the study and the treatment failures Outcome variables Total ORS consumed (ml) No. of stools 024 h 2548 h 4972 h Time of first semiformed stool (h) Time of first rehydration (h) Weight gain (g) Total IVF needed (ml) Duration of post intervention diarrhea (h) SB group Mean SD 3156.282109.92 [n=43] 24.7619.94 [n=43] 10.968.04 [n=25] 10.446.27 [n=9] 39.4823.09 [n=43] 5.943.16 [n=39] 409.65410.27 [n=42] 919.00466.25 [n=10] 49.5523.72 [n=43] Placebo group Mean SD 4262.662169 [n=47] 29.0020.00 [n=47] 16.2117.39 [n=37] 15.2714.01 [n=22] 54.1328.21 [n=46] 8.707.12 [n=40] 522.19561.8 [n=46] 1236.471013 [n=17] 66.3429.27 [n=47] P 0.016 0.318 0.164 0.332 0.009 0.030 0.285 0.362 0.004 95% CI for difference [(2004.4-(208.3)] (12.6-4.14) (12.71-2.2) (14.84-5.18) [25.5-(3.79)] [5.22-(0.27)] (320.2-95.17) (1021.7-386.8) [28.13-(5.43)]

Indian J Pediatr (April 2012) 79(4):478482

481

5.27 to 0.27) less in SB group (Table 3). Whereas, the mean post intervention duration of diarrhea and time of appearance to first semiformed stool continued to be significantly less in the SB group when the cases that left the study or became treatment failure were excluded. All the outcome variables were comparable between the two groups having rotavirus positivity in their stools. There were 8 (4 in each group) cases that were culture positive cholera. There has been no adverse event in the form of persistence of vomiting or deterioration in general clinical condition of the patients in either SB group or placebo group.

Discussion The present study found a significant difference in the duration of post intervention diarrhea between SB (52.85 24.6 h) vs. placebo group (64.6130.9 h). The other significant outcome variable was the time of appearance of first semi formed stool (39.4823.09 h in SB group versus 54.1328.21 h in placebo group). The difference in time of first rehydration and total ORS consumed between the two groups also became significant after excluding the cases that left the study or became treatment failure. The rotavirus cases were too few to give any worthwhile analysis. Of the studies [613] done to see the role of SB in acute childhood diarrhea, few have been done in the developing world [7, 911] where the whole burden of diarrheal illnesses exists. Among the two studies from Pakistan, one [7] found mean duration of the illness decreased (P 0.006) by a mean of 1.1 d (95% CI 1.4 0.24 d) and frequency and consistency of stools improved on day 3 in the SB group. The other [9] study also found a significant decrease in duration of diarrhea (3.6 vs. 4.8 d, P 0.001) and stool frequency in the SB group. Both these studies had used SB in the similar dosage as the present study and reported no adverse effects as the present study. More recently in a study [10] done in Argentinian children aged 324 mo, the authors found that SB reduces duration of diarrhea, total number of stools and number of watery stools passed on 4th and 7th day. In this study half the dose of SB as compared to the present study was given. Khin Htwe et al. [11] studied the efficacy of SB in acute diarrhea in hospitalized children aged 3 mo to 10 y. They found the mean duration of diarrhea was 3.08 d in the SB group and 4.68 d (P<0.05) in the control group. Some studies used very hetrogenous age group [9, 11] and few [9, 10] were also looking at the preventive role of SB in acute diarrhea. Hence the methodology of these studies was flawed with the primary outcome variable not being clear. In an earlier study [8] from Turkey, reduction in duration of diarrhea (4.72.5 vs. 5.53.2 d, P 0.03) and

hospital stay (2.9 vs. 3.9 d, P<0.001) was seen in the SB group. In a recent study [12] done in Turkish children, SB was as effective as yogurt fluid and on day 3, more children in SB had resolved diarrhea (48.5% vs. 25.5%). Lee Luyer et al. [13] found that when infants were given 26 mg of SB per 100 ml of formula milk, the beneficial effects were seen in the duration of diarrhea (35.4+/3.7 h vs. 67.1+/5 h; P <0.001) and daily weight gain (74.2+/26.4 g vs. 23.7+/6.7 g; P<0.05) in SB group vs. control. Albiet insignificant difference, there was more weight gain in placebo group (402 vs. 492 g). The plausible reasons could be more moderately malnourished children in the placebo group may have gained more weight on improvement in feeding, and since the placebo group received more intravenous fluids, hence over hydration could be another reason for more weight gain. Most of the earlier studies did not comment on adverse effects. The authors conclude that there is initial evidence available that SB may have a therapeutic role in the management of acute childhood diarrhea. Keeping in view the small sample size and the hospital based nature of the present study, the authors feel the results of the present study should be re-examined in a large community based trial before any recommendations are decided. But the significant reduction in the post intervention diarrheal duration and early improvement in stool consistency does raise hopes about the role of SB in management of acute childhood diarrhea as an adjunct to low osmolarity ORS, zinc and feeding.

Conflict of Interest None

Role of Funding Source None

References
1. Kosek M, Bern C, Guerrant RL. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull World Health Organ. 2003;81:197204. 2. O'Ryan M, Prado V, Pickering LK. A millennium update on pediatric diarrheal illness in the developing world. Semin Pediatr Infect Dis. 2005;16:12536. 3. Huang JS, Bousvaros A, Lee JW, Diaz A, Davidson EJ. Efficacy of probiotic use in acute diarrhoea in children: a meta-analysis. Dig Dis Sci. 2002;11:26253634. 4. Van Niel CW, Feudtner C, Garrison MM, Christakis DA. Lactobacillus therapy for acute infectious diarrhoea in children:a meta-analysis. Pediatrics. 2002;109:67884. 5. McFarland LV, Elmer GW, McFarland M. Meta-analysis of Probiotics for the prevention and treatment of Acute Pediatric Diarrhea. Int J Probiotics Prebiotics. 2006;1:6376. 6. Cetina-Sauri G, Sierra Basto G. Therapeutic evaluation of Saccharomyces boulardii in children with acute diarrhea. Annales De Pediatrie. 1994;41:397400.

482 7. Hafeez A, Tariq P, Ali S, Kundi ZU, Khan A, Hassan M. The efficacy of saccharomyces boulardii in the treatment of acute watery diarrhea of children: A multicentre randomized controlled trial. J Coll Physicians Surg Pak. 2002;12:4324. 8. Kurugol Z, Koturoglu G. Effect of Saccharomyces boulardii in children with acute diarrhoea. Acta Pediatr. 2005;94:447. 9. Billo AG, Memon MA, Khaskheli SA, et al. Role of probiotic (Saccharomics boulardii) in management and prevention of diarrhea. World J Gastroenterol. 2006;12:455760. 10. Villarruel G, Rubio DM, Lopez F, et al. Saccharomyces boulardii in acute childhood diarrhoea: a randomized, placebo-controlled study. Acta Pediatr. 2007;96:53841.

Indian J Pediatr (April 2012) 79(4):478482 11. Htwe K, Yee KS, Tin M, Vandenplas Y. Effect of Saccharomyces boulardii in the treatment of acute watery diarrhea in Myanmar children: A randomized controlled study. Am J Trop Med Hyg. 2008;78:2146. 12. Eren M, Dinleyici EC, Vandenplas Y. Clinical efficacy comparison of Saccharomyces boulardii and yogurt fluid in acute nonbloody diarrhea in children: a randomized, controlled, open label study. Am J Trop Med Hyg. 2010;82:48891. 13. Le Luyer B, Makhoul G, Duhamel JF. A multicentric study of a lactose free formula supplemented with Saccharomyces boulardii in children with acute diarrhea. Arch Pediatr. 2010;17:45965.