Triple therapy using a proton pump inhibitor with clarithromycin and amoxicillin or metronidazole given twice daily remains the recommended first-choice treatment for Helicobacter pylori infection (1,2). One major problem with this therapy is increased resistance to the antibiotics used in the triple therapy (3,4). In addition, adverse effects are commonly experienced by patients receiving H pylori eradication therapy and are reported by approximately 5% to 30% of patients receiving triple therapy (5). The most common side effect is gastrointestinal upset, manifested by diarrhea, nausea, or vomiting (5). Measures to overcome these problems include the use of probiotics, which are live microbial food ingredients that are beneficial to health. The rationale for the use of probiotics as adjunctive treatment for H pylori infection is based on the results of studies that have shown that various lactobacilli (eg, Lactobacillus johnsonii La1, L acidophilus CRL 639, L casei), or their metabolic products, can inhibit or kill H pylori in vitro (6,7).
A recent systematic review (5) evaluated the effects of probiotics on H pylori eradication rates and the side effects of anti–H pylori treatment. In that review, 14 randomized controlled trials (RCTs) of various methodological quality involving 1671 patients were identified. In patients with H pylori infection, probiotic supplementation improved eradication rates. In 2 RCTs that evaluated patients with eradication failure, probiotic supplementation also improved eradication rates. Probiotics reduced therapy-related side effects overall and reduced individual symptoms of diarrhea, epigastric pain, nausea, and taste disturbance.
Critics of using a meta-analytic approach to assess the efficacy of probiotics argue that the beneficial effects of probiotics seem to be strain specific; thus, pooling data on different strains may result in misleading conclusions. Furthermore, with only 1 exception (8), all trials included in the analysis were carried out in adults. Additionally, several experts and reviews have called for further studies to determine the effects of probiotic supplementation of H pylori eradication therapy and to identify the most effective strain(s). Given these considerations, we aimed to investigate in children the effectiveness of Lactobacillus GG (LGG) when administered concomitantly with anti–H pylori eradication therapy consisting of 2 antibiotics and a proton pump inhibitor. We initiated this study knowing that 3 previous trials with LGG had shown negative results (9–11). The rationale for conducting our trial was based on conflicting data about the effects of LGG in preventing antibiotic-associated diarrhea, which suggested different responses in adult (12) and pediatric populations (13,14) and that results observed in either population cannot be simply extrapolated to the other.
PATIENTS AND METHODS
We conducted a double-blind, randomized, placebo-controlled trial from February 2003 through June 2006. The institutional ethics committee approved the study. Written informed consent was obtained from the parent or guardian of each child included in the study.
We included children 5 to 17 years old who were hospitalized at the Department of Pediatric Gastroenterology and Nutrition with H pylori infection confirmed by 2 of 3 tests: the 13C-urea breath test (UBT), histopathology (hematoxylin and eosin staining), or the rapid urease test. We excluded children with acute gastrointestinal disease (eg, acute diarrhea), chronic gastrointestinal disease (eg, inflammatory bowel disease, celiac disease), renal and/or liver failure, obvious nervous system disease, and antibiotic and/or acid suppression and/or children who had received probiotic treatment in the previous 7 days.
Randomization and Blinding
We used computers to generate the allocation sequence and a randomization list. All of the study personnel, patients, and personnel involved in the conduct of the study were unaware of treatment assignments throughout the study. LGG and the control product were packed in identical forms. The study products were supplied by Dicofarm, Rome, Italy, who had no role in the conception, design, or conduct of the study or in the analysis or interpretation of the data. Randomization codes were secured until all of the data entry was complete.
The eligible children were randomly assigned to receive a 7-day, triple eradication therapy regimen consisting of 2 antibiotics (amoxicillin tablets, 25 mg/kg twice per day [maximum daily dose 1.5 g] and clarithromycin tablets, 10 mg/kg twice per day [maximum daily dose 1 g]) plus a proton pump inhibitor (omeprazole capsules, 0.5 mg/kg, twice per day) supplemented with LGG at a dose of 109 colony-forming units (CFU) or a comparable placebo (maltodextrine). Both the active treatment and the placebo were taken orally twice daily for the duration of the eradication therapy.
Each patient received forms/diaries to record the frequency of side effects (see below) and any symptoms they considered important. In total, 4 survey forms were filled out: 1 before the treatment; the second on the 7th day (the end of treatment); the third on the 21st day (2 weeks after treatment); and the fourth 4 to 6 weeks after treatment. H pylori infection was assessed at the beginning of the study and at 4 to 6 weeks after the completion of the intervention. The children were evaluated clinically at study entry, at day 7, at day 21, and at 4 to 6 weeks after enrollment. At the latter visit, the standard 13C-labeled UBT was performed by use of 75 mg 13C urea and orange juice as a test meal, as previously described (15–17). Baseline and 30-minute breath samples were analyzed by use of an isotope ratio mass spectrometer (ABCA, Europe Scientific Ltd, UK). The diagnostic cutoff value for the standard 13C-UBT was δ13CO2 ≥5.0% over the baseline (18). In the event of loose or watery stools during the treatment, patients were advised to contact their physicians and to bring stool samples for microbiological analysis to exclude an infectious origin of the diarrhea. Treatment compliance was assessed by direct interview of the patient (or parent) and a review of the diary cards (which documented the number of daily doses taken). Treatment responses also were assessed by evaluation of the diaries.
The primary outcome measure was the rate of H pylori eradication, which had to be confirmed by a negative 13C-UBT result at least 4 weeks after treatment. The secondary outcome measures were the proportion of patients with treatment-associated diarrhea, defined as 3 or more loose or watery stools per day for a minimum of 48 hours occurring during and/or up to 2 weeks after the end of the antibiotic therapy; any gastrointestinal side effects during anti–H pylori therapy, including abdominal pain, nausea, vomiting, constipation, flatulence, taste disturbance, or loss of appetite (only new symptoms not occurring before the treatment were considered); and the need for discontinuation of the anti–H pylori therapy.
Data were analyzed with computer software StatsDirect (version 2.5.6. [2006-04-15]; Iain E. Buchan). All statistical tests were 2-tailed and were performed at the 5% level of significance. The Student t test was used to compare means of continuous variables approximating a normal distribution. For non-normally distributed variables, the Mann-Whitney U test was used. The χ2 test or the Fisher exact test was used, as appropriate, to compare percentages. The same software was used to calculate the relative risk (RR), 95% confidence interval (CI), and number needed to treat. The difference between study groups was considered significant when the P value was <0.05 or when the 95% CI for RR did not exceed 1.0 (equivalent to P < 0.05). We report the analysis based on allocated treatment; that is, the cases of all of the participants in a trial for which outcome data were available were analyzed according to the intervention to which they were assigned, whether or not they received it. Inasmuch as all of the participants for whom outcome data were available sufficiently complied with the trial protocol, there was no difference between this and per-protocol analyses.
A potential problem with the analysis based on allocated treatment is that missing responses can lead to bias unless the absence of an observation is independent of outcome. Therefore, for both primary outcome measures, we also investigated the effect of various methods of handling missing responses in a trial (19). That is, we compared outcomes in both treatment groups (vs placebo) assuming all of the patients in both groups with an unknown outcome to have had either a good or a poor outcome, extreme case favoring of LGG, and extreme case favoring of placebo.
Figure 1 shows the flow diagram of the study participants' progression through the study. Of the 83 children who underwent randomization, 44 were assigned to the LGG group and 39 were assigned to the placebo group; 66 (80%) were included in the final analysis. Excluded from the analysis were 17 children for lack of diary and/or UBT. There was no significant difference in the dropout rate between the groups. Baseline demographic and clinical characteristics were similar in the LGG and placebo groups (Table 1).
H pylori Eradication
The outcome measures are summarized in Table 2. We found no significant difference between the study groups with respect to H pylori eradication rates. Of the 34 children in the LGG group, 23 (69%) experienced eradication compared with 22 of the 32 children (68%) in the placebo group (RR 0.98, 95% CI 0.7–1.4).
Investigation of the Potential Effects of Missing Responses
Assuming all patients in both groups with an unknown outcome to have had either a good or a poor outcome, our results demonstrate no significant difference in outcomes between the 2 treatments. Similar results were obtained by assuming extreme case favoring of placebo (ie, all the patients with an unknown outcome in the LGG group and none of the patients in the placebo group had H pylori eradication). The only exception was upon assuming (the unlikely) extreme case favoring of LGG (ie, all of the patients with an unknown outcome in the placebo group and none of the patients in the LGG group had H pylori eradication), which showed a significant difference between the groups (Table 3).
The risk of therapy-related diarrhea seemed to be lower in the LGG group than in the placebo group, but the difference was not statistically significant (6% vs 20%, RR 0.3, 95% CI 0.07–1.2). We found no significant difference between the study groups with respect to all adverse effects and any specific adverse effect. There was no need for discontinuation of the eradication treatment.
The results of this randomized, double-blind, placebo-controlled study demonstrate that LGG as an adjunct to standard triple eradication therapy (composed of amoxicillin, clarithromycin, and a proton pump inhibitor) was ineffective in increasing H pylori eradication rates and reducing therapy-related side effects.
Comparison with Previous Reports
With respect to H pylori eradication, the results of this trial are in line with observations in adults. Three previous RCTs (9–11) demonstrated that LGG compared with placebo administered during H pylori eradication therapy had no effect on eradication rates.
Interestingly, the present study, in contrast to all studies in adults with H pylori infection of which we are aware, revealed no evidence of a reduced incidence of side effects in the LGG-supplemented group. Several factors may contribute to explaining this difference. First, this study may have suffered from insufficient power to detect a significant difference between the 2 interventions. Second, at least when it comes to therapy-related diarrhea, the lack of difference observed between the 2 interventions may be due to the rather stringent definition that we used (≥3 loose or watery stools per day for a minimum of 48 hours). Such a definition allowed us to differentiate between clinically relevant conditions and clinically unimportant changes in the consistency of stools. However, the definitions of therapy-related diarrhea in published studies vary (or are not reported). Third, LGG may not have been administered at a sufficient dosage and/or duration to influence colonic bacterial flora. We chose a daily intake of 109 CFU twice daily, which exceeds the minimum dose of 109 CFU daily suggested in the literature for therapeutic purposes (20). It is noteworthy, however, that the optimal dosage and treatment duration of LGG therapy have not been clearly established. Fourth, theoretically, the viability of the LGG in the study product may have been a problem. This is unlikely, however, because the other study performed at the same time in our department with the study products provided by the same manufacturer showed a benefit of LGG (21). We abstained from stool analyses to assess colonization because the ability of LGG to survive transit through the gastric and biliary secretions and successfully colonize the colon has been well documented previously (22). Finally, conflicting data about the effects of LGG suggest a different response in adult and pediatric populations and that results observed in either population cannot be simply extrapolated to the other.
Comparison with Previous Reports in Children
In the pediatric population, to our knowledge, only 3 RCTs have evaluated whether consumption of probiotics could increase H pylori eradication rates and reduce the side effects of treatment. In the first RCT, (8) 86 children who had dyspepsia for more than 3 months and H pylori infection, diagnosed on the basis of 2 biopsy-based methods and a stool antigen test, were allocated to a 7-day course of eradication triple therapy with omeprazole, amoxicillin, and clarithromycin, or the same therapy supplemented with fermented milk containing Lactobacillus casei DN-114 001 (1010 CFU daily) for 14 days. Eradication rates were assessed by the H pylori stool antigen test and the 13C-UBT carried out 4 weeks after the end of treatment. H pylori infection was considered eradicated if the results of both tests were negative. Both per-protocol and intention-to-treat analyses were performed. The latter analysis demonstrated that patients in the probiotic-supplemented group had a greater eradication rate than did those in the placebo group (84.6% vs 57.5%, RR 1.47, 95% CI 1.1–2, number needed to treat 4, 95% CI 3–13). Similar to our findings, the incidence of side effects did not differ between groups. Drug compliance was good throughout the study. Perhaps the most exciting and clinically relevant finding of this trial is the increased eradication rate, in contrast to findings in our study and many studies in adults.
The second trial evaluated the effectiveness of a commercial yogurt containing Bifidobacterium animalis and L casei (107 CFU) as an adjuvant to antibiotic triple therapy for the eradication of H pylori infection in 65 children from Argentina. No significant difference in H pylori eradication rates at 1 month and 3 months between the treated group and the control group were found. Relative risks between groups were 0.87 (95% CI 0.58–1.32) in the first month and 0.97 (95% CI 0.64–1.46) in the third month. Side effects were not assessed (23).
In a recent randomized, double-blind RCT conducted in Italy, 40 H pylori–positive children (median age 12.3 years) treated with 10-day sequential therapy were randomly assigned to receive L reuteri ATCC 55730 (108 CFU) or placebo for 20 days (24). In the group of children receiving L reuteri, the Gastrointestinal Symptom Rating Score was lower than that in children in the control group during eradication therapy (4.1 ± 2 vs 6.2 ± 3, P < 0.01) and at the end of follow-up (3.2 ± 2 vs 5.8 ± 3.4, P < 0.009). However, there was no significant difference between groups in the eradication rates (17/20 vs 16/20). Given the results from these studies, it is evident that not all probiotics are created equal and that each strain must be evaluated individually.
Strengths and Limitations
We used adequate methods for the generation of the allocation sequence and allocation concealment. We made efforts to maintain blinding throughout selection, treatment, monitoring, data management, and data analyses. Data were obtained from 80% of the participants. These features minimize the potential for bias. The potential limitation of this trial is that we used paper forms to measure side effects. The validity of paper diary records is sometimes questioned (25). Well-known problems with paper diaries include poor adherence and retrospective (just before a visit) recording (25); however, limited resources did not allow us to use more reliable measures to record outcomes. A further limitation is that we assessed only the presence or absence of symptoms (eg, abdominal pain) and not the severity of the pain. Finally, owing to technical problems with the production and delivery of the study product, the trial was performed over a long period with possible changes in drug resistance and H pylori prevalence.
The results of this double-blind, placebo-controlled, randomized study showed that the use of LGG as an adjunct to conventional triple H pylori eradication therapy had no effect on eradication rates and therapy-related side effects. Whether other probiotic strains have such effects needs to be substantiated in further randomized trials.
The authors thank Dr Izabela Łazowska-Przeorek and Marcin Kalisz for their contribution to the study.
1. Gold BD, Colletti RB, Abbott M, et al
, North American Society for Pediatric Gastroenterology and Nutrition. Helicobacter pylori
infection in children: recommendations for diagnosis and treatment. J Pediatr Gastroenterol Nutr 2000; 31:490–497.
2. Bourke B, Ceponis P, Chiba N, et al
. Canadian Helicobacter Study Group Consensus Conference: update on the approach to Helicobacter pylori infection in children and adolescents: an evidence-based evaluation. Can J Gastroenterol 2005; 19:399–408.
3. Koletzko S, Richy F, Bontems P, et al
. Prospective multicentre study on antibiotic resistance of Helicobacter pylori
strains obtained from children living in Europe. Gut 2006; 55:1711–1716.
4. Oderda G, Shcherbakov P, Bontems P, et al
. European Pediatric Task Force on Helicobacter pylori
: results from the pediatric European register for treatment of Helicobacter pylori
(PERTH). Helicobacter 2007; 12:150–156.
5. Tong JL, Ran ZH, Shen J, et al
. Meta-analysis: the effect of supplementation with probiotics
on eradication rates and side effects during Helicobacter pylori
eradication therapy. Aliment Pharmacol Ther 2007; 25:155–168.
6. Bhatia SJ, Kochar N, Abraham P, et al
. Lactobacillus acidophilus
inhibits growth of Campylobacter pylori
in vitro. J Clin Microbiol 1989; 27:2328–2330.
7. Bernet MF, Brassart D, Neeser JR, et al
. Lactobacillus acidophilus
LA 1 binds to cultured human intestinal cell lines and inhibits cell attachment and cell invasion by enterovirulent bacteria. Gut 1994; 35:483–489.
8. Sykora J, Valeckova K, Amlerova J, et al
. Effects of a specially designed fermented milk product containing probiotic Lactobacillus casei
DN-114 001 and the eradication of H. pylori
in children: a prospective randomized double-blind study. J Clin Gastroenterol 2005; 39:692–698.
9. Armuzzi A, Cremonini F, Ojetti V, et al
. Effect of Lactobacillus GG supplementation on antibiotic-associated gastrointestinal side effects during Helicobacter pylori
eradication therapy: a pilot study. Digestion 2001; 63:1–7.
10. Armuzzi A, Cremonini F, Bartolozzi F, et al
. The effect of oral administration of Lactobacillus GG on antibiotic-associated gastrointestinal side-effects during Helicobacter pylori
eradication therapy. Aliment Pharmacol Ther 2001; 15:163–169.
11. Cremonini F, Di Caro S, Covino M, et al
. Effect of different probiotic preparations on anti-Helicobacter pylori
therapy-related side effects: a parallel group, triple blind, placebo-controlled study. Am J Gastroenterol 2002; 97:2744–2749.
12. Thomas MR, Litin SC, Osmon DR, et al
. Lack of effect of Lactobacillus
GG on antibiotic-associated diarrhea: a randomized, placebo controlled trial. Mayo Clin Proc 2001; 76:883–889.
13. Vanderhoof JA, Whitney DB, Antonson DL, et al
GG in the prevention of antibiotic-associated diarrhea in children. J Pediatr 1999; 135:564–568.
14. Arvola T, Laiho K, Torkkeli S, et al
. Prophylactic Lactobacillus
GG reduces antibiotic-associated diarrhea in children with respiratory infections: a randomized study. Pediatrics 1999; 104:e64.
15. Rowland M, Lambert I, Gormally S, et al
. Carbon 13-labeled urea breath test for the diagnosis of Helicobacter pylori
infection in children. J Pediatr 1997; 131:815–820.
16. Bode G, Rothenbacher D, Brenner H, et al
. Variation in the 13Curea breath test value by nationality in Helicobacter pylori
-infected children. Scand J Gastroenterol 1998; 33:468–472.
17. Ciok J, Tacikowski T, Dzieniszewski J. Urea breath test (UBT) in the diagnosis of Helicobacter pylori
infection [in Polish]. Gastroenterologia 2003; 10:241–246.
18. Cave DR, Zanten SV, Carter E, et al
. A multicentre evaluation of the laser assisted ratio analyser (LARA): a novel device for measurement of 13CO2 in the 13C-urea breath test for the detection of Helicobacter pylori
infection. Aliment Pharmacol Ther 1999; 13:747–752.
19. Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomized controlled trials. BMJ 1999; 319:670–674.
20. Lee YK, Salminen S. The coming age of probiotics
. Trends Food Sci Technol 1995; 6:241–245.
21. Gawrońska A, Dziechciarz P, Horvath A, et al
. A randomized double-blind placebo-controlled trial of Lactobacillus GG for abdominal pain disorders in children. Aliment Pharmacol Ther 2007; 25:177–184.
22. Goldin BR, Gorbach SL, Saxelin M, et al
. Survival of Lactobacillus species (strain GG) in human gastrointestinal tract. Dig Dis Sci 1992; 37:121–128.
23. Goldman CG, Barrado DA, Balcarce N, et al
. Effect of a probiotic food as an adjuvant to triple therapy for eradication of Helicobacter pylori
infection in children. Nutrition 2006; 22:984–988.
24. Lionetti E, Miniello VL, Castellaneta SP, et al
. Lactobacillus reuteri therapy to reduce side-effects during anti-Helicobacter pylori
treatment in children: a randomized placebo controlled trial. Aliment Pharmacol Ther 2006; 24:1461–1468.
25. Stone AA, Shiffman S, Schwartz JE, et al
. Patient non-compliance with paper diaries. Br Med J 2002; 324:1193–1194.