Bacillus clausii therapy to reduce side-effects of anti-helicobacter pylori treatment: randomized, double-blind, placebo controlled trial

Aliment Pharmacol Ther 2004; 20: 1181–1188.
Bacillus clausii therapy to reduce side-effects of anti-Helicobacterpylori treatment: randomized, double-blind, placebo controlled trial E . C . N I S T A * , M . C A N D E L L I * , F . C R E M O N I N I * , I . A . C A Z Z A T O * , M . A . Z O C C O * , F . F R A N C E S C H I * ,G . C A M M A R O T A * , G . G A S B A R R I N I * & A G A S B A R R I N I  Departments of *Internal Medicine and  Medical Pathology, Catholic University, Rome, Italy Accepted for publication 27 September 2004 treatment. (ii) The same 7 days triple therapy and placebo t.d.s. for 14 days starting from the first day of Background: Helicobacter pylori eradication fails in treatment. Side-effects were assessed using a validated about 10% of patients, particularly because of the questionnaire and were recorded for 4 weeks from the occurrence of resistance to antibiotics and side-effects.
During anti-H. pylori therapy, probiotics have been Results: The incidences of nausea, diarrhoea and successfully used to reduce the incidence of side- epigastric pain in patients treated with B. clausii were significantly lower than in placebo group, in both PP Aim: To evaluate the effect of Bacillus clausii, a probiotic, and ITT analysis. Equally, intensity of nausea and on incidence (primary variable) and severity of antibi- diarrhoea in patients treated with B. clausii was otic-associated side-effects during anti-H. pylori therapy.
significantly lower than in placebo group. There were Methods: One hundred and twenty H. pylori-positive no differences in adherence to treatment and H. pylori patients were randomly screened to receive: (i) a standard 7 days triple therapy with rabeprazole 20 mg Conclusion: In symptom-free, H. pylori-positive subjects b.d., clarithromycin 500 mg b.d. and amoxicillin 1 g B. clausii bacteriotherapy reduces the incidence of the b.d. and B. clausii t.d.s. (each preparation containing most common side-effects related to anti-H. pylori 2 · 109 spores) for 14 days starting from the first day of antibiotic therapy compared with placebo.
ation and failure to eradicate H. pylori. Poor compliance to treatment is one of the determinants of failure to Therapeutic schemes to eradicate Helicobacter pylori eradicate H. pylori. Despite the number of dropouts is infection are based on the association of antibiotics and relatively small in clinical trials, this may not be the case a proton pump inhibitor.1–8 Treatment outcome depends on the class of antibiotic administered, dosages used, Probiotics are defined as microbial cell preparations or therapy duration, bacterial resistance and patients’ components of microbial cells that can beneficially impact compliance. Gastrointestinal side-effects during antibi- human health.9 The use of probiotics has recently been otic therapy include diarrhoea, nausea, taste distortion, proposed to increase patients’ tolerability by limiting side- stomatitis and bloating. These side-effects reduce treat- effects of anti-H. pylori eradicating therapies.10–12 ment tolerability and may cause treatment discontinu- The probiotic strains previously used in H. pylori eradication trials include Lactobacillus GG, Saccharomy- Correspondence to: Prof. A. Gasbarrini, Department of Internal Medicine, ces boulardii, or combinations of different strains. In Catholic University, Rome, Italy.
E-mail: [email protected] these studies, the addition of probiotics improved the tolerability of eradication regimens, with limited effect – Sixty patients (male/female 33/27, mean age 46.2 ± 12.83) were randomly assigned to receive a triple Bacillus clausii is a probiotic widely used in Italy since the therapy based on clarithromycin 500 mg b.d., 1960s for viral diarrhoea in children and for antibiotic amoxicillin 1 g b.d., rabeprazole 20 mg b.d. for related side-effects.13 B. clausii spores can survive the 7 days plus a probiotic preparation, one vial t.d.s.
gastric pH, activate and reach the intestinal tract where (each vial containing 2 · 109 spores of B. clausii, they germinate to vegetative forms.14, 15 Experimental Enterogermina, Sanofi–Synthelabo OTC, Milan, Italy) data suggest that both B. clausii spores and cells can for 14 days, during eradication therapy and 1 week adhere to the bowel wall and colonize the mucosa.16 Our aim was to assess, in a double-blind, randomized, – Sixty patients (male/female 25/35, mean age 43.1 ± placebo-controlled trial the effect of oral bacteriotherapy 13.36) were randomly assigned to a placebo pre- with B. clausii on gastrointestinal side-effects occurring paration, one placebo vial t.d.s., during the 7-day eradication therapy and for 1 week thereafter. Theplacebo preparation, looked identical in colour, size,shape, weight and taste to the probiotic preparation.
The study was a single centre, double-blind, prospective, Each patient was required to complete a validated daily randomized, placebo controlled study performed at the diary for 4 weeks, starting from the first day of the Gastroenterology and Internal Medicine Day Hospital of eradicating treatment.11 The diary contains a question- Gemelli Hospital of Rome, Italy. One hundred and twenty naire (slightly modified from DeBoer et al.) 17 evaluating consecutive H. pylori-positive patients free from gastro- onset, intensity and frequency of gastrointestinal side- intestinal symptoms (spouses or relatives of patients with effects: taste distortion, loss of appetite, nausea, vomit- H. pylori associated gastrointestinal diseases who asked to ing, epigastric pain, bloating, diarrhoea, constipation be tested and treated for H. pylori infection) were enrolled and skin rash. The intensity of symptoms was rated using a scale, in which 0, 1, 2 and 3, respectively Patients were considered eligible to enter the study if corresponded to absent, mild, moderate and severe they were between 18 and 65 years old, free of symptoms. An overall judgment of tolerability was gastrointestinal symptoms in the previous 3 months assessed by the patient at the end of both the first and and affected by gastric H. pylori infection as confirmed second weeks of treatment. Treatment tolerability was by a 13C-urea breath test. Exclusion criteria were: recent scored with a scale in which 1, 2, 3, 4 and 5 (within the previous 3 months) use of anti-microbial respectively corresponded to 1, no side-effect observed; agents, bismuth compounds, proton pump inhibitors 2, mild side-effect(s), non-interfering with daily activit- and H2 receptor antagonists, laxatives, anti-diarrhoeals, ies; 3, moderate side-effect(s), slightly interfering with other probiotic preparations, alcohol or illicit drug daily activities; 4, severe side-effect(s), interfering with abuse. Patients with acute or chronic gastrointestinal daily activities but not leading to treatment interruption diseases, or with major concomitant diseases including and 5, severe side-effect(s), producing treatment inter- psychiatric disorders and pregnant or lactating women ruption. Adherence to treatment was evaluated by were also excluded from the study. Patients under counting the vials returned by the subject [patients who chronic drug treatment were considered suitable if they returned <80% of empty vials were not included in the had been receiving such treatments for >3 months. All ‘per protocol population’ (PP) analysis].
patients signed a written informed consent.
Helicobacter pylori eradication was evaluated by means Using a permuted blocks randomization (1:1), 120 of a 13C-urea breath 6 weeks after the end of the patients were assigned to one of the following parallel treatment. A delta 13C over baseline value higher than 3.5 was considered positive for active H. pylori infection.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 B . C L A U S I I B A C T E R I O T H E R AP Y A N D S I D E - E F F E C T S O F H . P Y L O R I T H E R A P Y Table 1. Demographic data of the patients included in the safetypopulation (n ¼ 114) For statistical analysis the following population wereconsidered: patients who have taken at least one dose of study – ‘Intention-to-treat population’ (ITT), defined as all randomized patients who have taken at least one in age and baseline symptom scores (all patients were dose of study treatment and have returned the first symptom free at enrolments) between the placebo diary, reporting at least one symptom assessment and the B. clausii groups. On the contrary, a higher during the 7-day eradication therapy. Missing data prevalence of male gender was observed in B. clausii because of premature discontinuation of treatment group and female gender in placebo group.
were considered treatment failures for the analysisof symptom incidence and were replaced using Trial flow. Six patients (three in the B. clausii group, the mean of the available daily scores of the three in the placebo group) did not start the assigned patient for the analysis of symptoms intensity and treatment. Eight patients (three in the B. clausii group, five in the placebo group) did not return the first diary.
– ‘Per protocol population’, defined as all randomized Further six patients (four in the B. clausii group, two in patients who completed the study without protocol the placebo group) were not included in the per- protocol analysis because they did not return the Descriptive analysis of demographic, anamnestic and second diary or because of withdrawal or poor clinical data by treatment was performed. Dichotomous compliance (i.e. <80% of the vials were recovered) variables were analysed by v2 test and relative risk (RR) and its 95% confidence interval (CI) were calculated.
The H. pylori eradication rate was similar between The differences between the two groups in symptoms B. clausii and placebo groups. In particular, ITT analysis intensity, summarized in the overall mean over time, has shown H. pylori was eradicated in 39 of 54 patients were assessed using analysis of variance, after rank (72.2%) in the B. clausii group and in 37 of 52 patients transformation of the summary measure, considering (71.15%) in the placebo group. In PP population, treatment and diary mean factors and treatment by H. pylori was eradicated in 39 of 50 patients (78%) in time interaction. For the overall judgements of tolerab- the B. clausii group and in 37 of 50 patients (74%) in ility, the distribution of patients with side-effects in the two treatment groups was evaluated by Kruskal–Wallistest. Tests were two sided and P-value <0.05 was Effect of Bacillus clausii on the incidence and intensity of considered statistically significant.
side-effects. Both ITT and PP analysis showed a signifi-cant difference between the two treatments in theincidence of nausea, diarrhoea and epigastric pain. In the ITT analysis, RR of occurrence of nausea were halved Based on a previous study,9 a total sample size of 120 in patients treated with B. clausii compared with placebo was calculated as adequate to detect, with an 80% after one (RR ¼ 0.5; 95% CI 0.31–0.88) and 2 weeks power and a two sided 5% significance level, a difference (RR ¼ 0.45; 95% CI 0.21–0.96) of bacteriotherapy. A between the two groups ‡20% in occurrence of nausea, greater reduction in the risk of diarrhoea was observed in the B. clausii group compared with the placebo groupafter one (RR ¼ 0.30; 95% CI 0.12–0.76) and 2 weeks(RR ¼ 0.38; 95% CI 0.08–1.9). For epigastric pain, after 1 week, the relative risk was 0.68, 95% CI 0.48–0.97 Baseline characteristics. Participant characteristics are (Table 2). The incidence of vomiting, constipation and showed in Table 1. There were no significant differences skin rash were also lower in the B. clausii group Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 Randomized
n = 120
Safety population
n = 114
ITT population
n = 106
PP population
n = 100
compared with the placebo group, although the differ- The individual patients’ overall assessment of tolerab- ences were not statistically significant.
ility in ITT population was better in the group treated The mean intensities and frequencies of nausea and with B. clausii than in the placebo group. The difference diarrhoea were also significantly lower in the B. clausii was statistically significant after 2 weeks of treatment Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 B . C L A U S I I B A C T E R I O T H E R AP Y A N D S I D E - E F F E C T S O F H . P Y L O R I T H E R A P Y Table 2. Incidence of symptoms during treatment. Data are presented as percentages and relative risks with 95% confidence intervals RR, relative risks; CI, confidence intervals; PP, per protocol population; ITT, intention-to-treat population.
* P < 0.05, **P < 0.01.
probiotic treatments during anti-H. pylori regimens (PPI, clarithromycin and tinidazole) were able to reduce In the present study, we report that patients using the incidence diarrhoea, nausea and taste distortion. In B. clausii during H. pylori eradication therapy experienced the current study, amoxicillin was used instead of a lower incidence of side-effects than subjects whose tinidazole for the first time based on the Maastricht regimen was supplemented with placebo. The symptoms 2–2000 Consensus guidelines.6 The different antibiotic with a lower incidence in the group of patients treated administered could explain the low incidence of taste with B. clausii compared with those supplemented with distortion experienced by patients in our study as taste placebo were diarrhoea, nausea and epigastric pain.
distortion is one of the commonest side-effects related to The differences were more marked in the first week of use of tinidazole and other nitroimidazoles.
treatment when antibiotics were given and when the Antibiotic related side-effects are common and usually incidence of side-effects was higher in both groups. The affect the gastrointestinal system. The bowel milieu is incidence of nausea, diarrhoea and epigastric pain was characterized by a high bacterial concentration, up to affected by B. clausii treatment, in particular with regard 1012–1014 CFU/mL in the colon; such bacteria coexist to the intensity and frequency of diarrhoea as it was in equilibrium with colonic mucosal cells. Antibacterial significantly lower in the probiotic treated group than drugs can alter this equilibrium causing potentially the placebo group. In previous studies,9, 11 oral pathogen species to prevail over the normal resident Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 Table 3. Summary measure of symptoms intensity and frequency PP, per protocol population; ITT, intention-to-treat population.
Table 4. Overall judgment on tolerability by time and treatment microflora. Modifications in the composition of the (intention-to-treat population population) intestinal bacteria may be induced by any antibiotic, butthe broad spectrum antibiotics (such as tetracyclines, aminoglycosides, macrolides and penicillins) are con- sidered the most frequent drugs responsible for these gastrointestinal side-effects. Moreover, macrolide anti- biotics, such as clarithromycin, have been related to increased contractility of gastrointestinal smooth mus- cle, which may lead to increased motility and acceler- Probiotics can also modify gut microflora. The probi- otic species most commonly used include spore formers, lactic acid bacteria and yeasts. Probiotics have been shown to be useful and effective in a number ofgastrointestinal diseases such as antibiotic-associated Kruskal–Wallis test: P ¼ 0.07 during the first week, P < 0.05 during diarrhoea19 and inflammatory bowel diseases.20 The the second week.
* For two patients data are not available.
explanations for the effects of probiotics in human Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 B . C L A U S I I B A C T E R I O T H E R AP Y A N D S I D E - E F F E C T S O F H . P Y L O R I T H E R A P Y disease include the synthesis of anti-microbial sub- We anticipate that in dyspeptic patients the supplemen- stances, the competition with pathogenic microorgan- tation with B. clausii will reduce the burden of side- isms for nutrients and microbial adhesion sites, the effects to an extent that is more clinically relevant, modification of toxins or toxin receptors, the incomplete however, it must be acknowledged that these data were lactose digestion and immune system modulation.21 not collected to prove this hypothesis. Thus an addi- The use of B. clausii as a probiotic species has been tional clinical study would be needed to prove this point.
based on more over than 40 years of clinical usage in It should also be noted that we did not observe any Italy with excellent tolerability and no report of side- difference between two groups in the H. pylori eradica- effects. Moreover it has several unique properties such tion rate. This result is in agreement with previous as the resistance to commonly used antibiotics, 22 the studies with living probiotic species, 9, 11 and suggest absence in normal resident intestinal flora and the that B. clausii in the form and concentration we sporogenic activity.13, 23–25 Bacillus clausii spores can administered does not possess antibacterial activity survive in the acid gastric environment, activate and against H. pylori. Of relevance, in both arms we attained reach the intestinal tract where they germinate to eradication rates lower than in previous studies using vegetative forms.14, 15 Yet, the actual mechanism of the same antibiotic regimen.26 These results were action of B. clausii spores in the restoration of intestinal unexpected and might be partly attributable to the flora has not been fully clarified. Experimental data rising antibiotic resistance in our geographical area.
suggest that both B. clausii spores and cells can adhere Resistance to clarithromycinin in Italy has been reported to the bowel wall, allowing mucosal colonization.16 A to be around 10%.27 Similar to previous experience 9, 11 previous study showed a positive action of the probiotic overall compliance of patients was not affected by in case of rotavirus diarrhoea and in children with B. clausii. However, the sample size was chosen to antibiotic-related gastrointestinal side-effects.13 evaluate difference in incidence of side-effects and not in This study however has some limitations. First, the eradication rate and overall compliance. Moreover, occurrence of side-effects in anti-H. pylori therapy is patients enrolled in this study asked to be eradicated.
mostly attributed to the use of antibiotics in moderate to Thus, they might have been particularly motivated to high doses and in combination. Although some specific complete the treatment independently on the incidence side-effects such as diarrhoea could be related to the of side-effects. Although eradication rate and overall disruption of gut microflora by antibiotics, the link with compliance are not affected by B. clausii supplementa- bowel microecology for other common side-effects such tion, it is still be an advantage to reduce side-effects. On a as nausea and epigastric pain is unproven. Until the case-by-case basis, the clinician should determine whe- mechanistic bases of the actions of probiotics are fully ther it is worth supplementing the patient with B. clausii understood, caution should be used in attributing to prevent mild to moderate side-effects, without affect- symptomatic benefits to probiotic oral therapy. Sec- ing compliance. Cost consideration and the preference of ondly, we did not perform stool assessment for bacterial the individual patient should also be taken into account.
recovery. However, faecal recovery studies have previ- In conclusion, this study shows that B. clausii treat- ously confirmed presence of B. clausii in the stools after ment during and after a standard seven day anti- oral administration of commercially available prepara- H. pylori regimen is associated with lower incidence of tions and B. clausii has been shown to be resistant to the self-reported side-effects and with a better tolerability to antibiotics used in the present study. Thirdly, the results multiple antibiotic treatment when compared with from our study can be generalized only to subjects who undergo antibiotic therapy without having gastrointes-tinal symptoms. In fact, we enrolled symptom-free subjects who wished to eradicate their H. pyloriinfection, in order to assess newly occurring therapy- 1 Basset C, Holton J, Ricci C, et al. Review article: diagnosis and related side-effects. The inclusion of symptom-free treatment of Helicobacter: a 2002 updated review. AlimenPharmacol Ther 2003; 17: 89–97.
subjects allowed for clearer interpretation of the effect 2 Di Caro S, Assunta Zocco M, Cremonini F, et al. Levofloxacin of supplementation with the probiotic on side-effects based regimens for the eradication of Helicobacter pylori. Eur J occurrence, but limited the effect size on symptom Gastroenterol Hepatol 2002; 14: 1309–12.
intensity because the baseline scores were close to zero.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188 3 Gene E, Calvet X, Azagra R, Gisbert JP. Triple vs. quadruple 15 Urdaci MC, Bressollier P, Pinchuk I, Bacillus clausii probiotic therapy for treating Helicobacter pylori infection: a meta- strains: antimicrobial and immunomodulatory activities. J analysis. Alimen Pharmacol Ther 2003; 17: 1137–43.
Clin Gastroenterol 2004; 38: S86–90.
4 Katelaris PH, Forbes GM, Talley NJ, Crotty B. A randomized 16 Angioi A, Zanetti S, Sanna A. Adhesiveness of Bacillus subtilis comparison of quadruple and triple therapies for Helicobacter strains to epithelial cells cultured in vitro. Microbial Ecology in pylori eradication: the quadrate study. Gastroenterology Health and Disease 1995; 8: 71–77.
17 de Boer WA, Thys JC, Borody TJ, Graham DY, O’Morain C, 5 Laine L, Hunt R, El-Zimaity H, Nguyen B, Osato M, Spenard J.
Tytgat GN. Proposal for use of a standard side effect scoring Bismuth-based quadruple therapy using a single capsule of system in studies exploring Helicobacter pylori treatment regi- bismuth biskalcitrate, metronidazole, and tetracycline given mens. Eur J Gastroenterol Hepatol 1996; 8: 641–3.
18 Peters DH, Clissold SP. Clarithromycin. A review of its anti- clarithromycin for eradication of Helicobacter pylori in duode- microbial activity, pharmacokinetic properties and therapeu- nal ulcer patients: a prospective, randomized, multicenter, tic potential. Drugs 1992; 44: 117–64.
North American trial. Am J Gastroenterol 2003; 98: 562–7.
19 Cremonini F, DiCaro S, Nista EC, et al. Meta-analysis: the effect 6 Malfertheiner P, Megraud F, O’Morain C, et al. Current con- of probiotic administration on antibiotic-associated diarrhoea.
cepts in the management of Helicobacter pylori infection – the Alimen Pharmacol Ther 2002; 16: 1461–7.
Maastricht 2–2000 Consensus Report. Alimen Pharmacol 20 Gionchetti P, Amadini C, Rizzello F, et al. Probiotics–role in inflammatory bowel disease. Digestive & Liver Disease 2002; 7 Mascitelli L, Pezzetta F. Quadruple treatments for Helicobacter pylori (comment). Lancet 2003; 361: 86.
21 Isolauri E. Probiotics in human disease. Am J Clin Nutr 2001; 8 Zullo A, Vaira D, Vakil N, et al. High eradication rates of Helicobacter pylori with a new sequential treatment (erratum 22 Bozdogan B, Galopin S, Gerbaud G, Courvalin P, Leclercq R.
appears in Aliment Pharmacol Ther 2003 May 1;17:1205).
Chromosomal aadD2 encodes an aminoglycoside nucleot- Alimen Pharmacol Ther 2003; 17: 719–26.
idyltransferase in Bacillus clausii. Antimicrob Agents Chemo- 9 Armuzzi A, Cremonini F, Bartolozzi F, et al. The effect of oral administration of Lactobacillus GG on antibiotic-associated 23 Senesi S, Celandroni F, Tavanti A, Ghelardi E. Molecular gastrointestinal side-effects during Helicobacter pylori eradica- characterization and identification of Bacillus clausii Strains tion therapy. Alimen Pharmacol Ther 2001; 15: 163–9.
marketed for use in oral bacteriotherapy. Applied & Environ- 10 Cremonini F, Canducci F, Di Caro S, et al. Helicobacter pylori mental Microbiology 2001; 67: 834–9.
treatment: a role for probiotics? Dig Dis 2001; 19: 144–7.
24 Mazza P, Zani F, Martelli P. Studies on the antibiotic resistance 11 Cremonini F, Di Caro S, Covino M, et al. Effect of differ- of Bacillus subtilis strains used in oral bacteriotherapy. Boll ent probiotic preparations on anti-Helicobacter pylori ther- apy-related side effects: a parallel group, triple blind, 25 Ciffo F. Determination of the spectrum of antibiotic resistance placebo-controlled study. Am J Gastroenterol 2002; 97: of the ‘Bacillus subtilis’ strains of Enterogermina. Chemioter- 12 Fuller R. Probiotics in human medicine. Gut 1991; 32: 439– 26 Danese S, Armuzzi A, Romano A, et al. Efficacy and tolerab- ility of antibiotics in patients undergoing H. pylori eradication.
13 Benoni G, Marcer V, Cuzzolin L, Raimo F. Antibiotic admin- Hepatogastroenterology 2001; 48: 465–7.
istration and oral bacterial therapy in infants. Chemioterapia 27 Franzin L, Pennazio L, Cabodi D, Rossigni F, Gioannini P.
Clarithromycin and amoxicillin susceptibility of Helicobacter 14 Ciffo F, Da Carro C, Giovannetti M. Gastric resistance of pylori strains isolated from adults with gastric or duodenal Bacillus subtilis spores used in oral bacteriotherapy: in vitro ulcer in Italy. Curr Microbiol 2000; 40: 96–100.
studies. Farmaci e terapia 1987; 3: 163–169.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 1181–1188

Source: http://www.bellybiotic.net/images/study1.pdf