| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Division of Gastroenterology, Department of Internal Medicine, Hyogo Medical College, Mukogawa-cho 1-1, Nishinomiya, Hyogo 663-8501, Japan 2Division of Gastroenterology, Department of Internal Medicine, Juntendo University, School of Medicine, Tokyo, Japan
Correspondence Hiroto Miwa miwahgi{at}hyo-med.ac.jp
Received July 9, 2004
Accepted October 28, 2004
Eradication treatment for Helicobacter pylori is known to cause mild but relatively frequent adverse effects. Some adverse effects such as diarrhoea and soft stools are related to disruption of the composition of the intestinal microflora. This study investigated the microfloral changes resulting from administration of an eradication regimen using proton pump inhibitor (PPI), amoxycillin and clarithromycin. Twenty-eight laboratory-bred Japanese macaques either were administered eradication treatment by this regimen for 7 days or received no medication. Faecal samples were collected for analysis on days 0, 8 and 15, and both aerobic and anaerobic cultures were performed. Among aerobic bacteria, Streptococcus had significantly decreased by day 8, while Enterococcus and Enterobacteriaceae had significantly increased. However, the total number of aerobic bacteria was not significantly decreased from pretreatment levels 1 day after completion of treatment. The number of anaerobic bacteria did not change significantly by day 8. However, the number of Lactobacillus and the detection rate of Bifidobacterium, Peptostreptococcus and Veillonella significantly decreased by day 8, although the number of Bifidobacterium, Peptostreptococcus and Veillonella had almost recovered up to the pretreatment levels 1 week after completion of treatment (day 15). These results suggest that the alterations in the composition of the intestinal microflora caused by the antimicrobial regimen that excludes metronidazole are different from those caused by the regimen including this drug. However, the alterations in bacterial microflora had almost reversed 7 days after completion of treatment in these macaques, which supports clinical findings that diarrhoea or soft stools in humans resolve relatively quickly after a similar treatment.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Analysis of profiles of the intestinal microflora would be useful in exploring the mechanisms of the diarrhoea associated with this treatment. A few reports have described microfloral changes during and after eradication treatment by regimens containing PPI and metronidazole (Adamsson et al., 1999; Bühling et al., 2001). However, publications focusing on microfloral changes brought about by the combination regimen of PPI, amoxycillin and clarithromycin treatment, which is the only regimen approved by the Japanese social security foundation (Asaka et al., 2001), are not available.
On the other hand, it is not easy to investigate changes in intestinal microflora after eradication treatment in humans due to methodological limitations such as difficult sampling procedures and the influence of food or environmental factors on intestinal microflora. Such methodological difficulties could be overcome using experimental animal models. Therefore, in this study, we investigated changes in intestinal microflora resulting from administration of an eradication regimen using PPI, amoxycillin and clarithromycin in Japanese macaques.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Drug administration and stool sampling.
Twenty animals were given the eradication treatment by the PPI/amoxycillin/clarithromycin regimen for 7 days, and eight macaques received no treatment (control animals). Briefly, after animals had fasted more than 12 h (no water restriction), a capsule containing 30 mg of lansoprazole, six capsules containing 1500 mg of amoxycillin and two tablets containing of 400 mg clarithromycin were orally administered to anaesthetized macaques once daily for 7 days (treatment group). Preparation for anaesthetization was done with intramuscular ketamine hydrochloride designed for animal use (Sankyo) at a dose of 10 mg kg–1 (Fukuda et al., 1992). After dosing, the anaesthetized macaques were endoscoped by a sterilized fibrescope GIF-P10 (Olympus Optical) to confirm that these drugs were in the stomach (Fukuda et al., 1992).
In both the treatment and control groups, stool samples were collected before (day 0), 1 day after (day 8) and 1 week after the dosing (day 15). Faeces of each macaque were collected directly from the rectum by a sterilized stick of our making.
Analysis of the intestinal microflora profiles.
Immediately after being obtained, stools were packed in a sterile vinyl bag and mixed well to attain a homogeneous state. Part of the stool (about 2 g) was taken from the centre portion of the mixed faeces, placed in a transport tube (SEEDTUBE Eiken, Eiken Chemical Co.) and kept at 4 °C.
The stool samples were transported to the hospital laboratory and were immediately processed for stool culture according to the method of Mitsuoka (1980). Briefly, an accurately weighed 1 g portion of faecal specimen was mixed well and diluted 10-fold in 9 ml of an anaerobic diluent. The composition of a 990 ml quantity of the anaerobic diluent was as follows: 37.5 ml salt diluent 1 (0.78 % K2HPO3), 37.5 ml salt diluent 2 (0.47 % KH2PO4, 1.18 % NaCl, 1.20 % (NH4)2SO4, 0.12 % CaCl2, 0.25 % MgSO4.H2O), 1 ml 0.1 % Resazurin diluent, 0.5 g L-cysteine.HCl-H2O, 2 ml 25 % L-ascorbic acid diluent, 50 ml 8 % Na2CO3 diluent, 0.5 g agar and 860 ml water. The suspension was diluted serially by 10-fold with the same diluent up to 108 dilution.
Aliquots (0.05 ml) of each diluted sample were inoculated on 15 different appropriate agar media, which are shown in Table 1, and incubated under the respective conditions required for optimal growth of each organism. All aerobic plates were incubated at 35 °C for 2–3 days and all anaerobic plates were incubated in an anaerobic chamber (Mitsubishi Gas Chemical Co.) at 35 °C for 3–5 days. After incubation, different colony types were counted and identified to the genus level on the basis of Gram staining, general morphology, biochemical tests and a growth test under aerobic conditions utilizing methods described elsewhere (Murray et al., 1999). The concentration was defined as the number of colony forming units (c.f.u.) per gram (g) stool sample. The detection limit was 102 c.f.u. (g stool)–1.
|
Physical status of the macaques during the experiment.
The general condition (facial expression, appearance and behaviour) and stool condition (diarrhoea or soft stool) of the animals were observed and recorded throughout the experimental period. Body weights were measured weekly under anaesthesia.
Statistics.
Student's t-test, Fisher's exact test and one-way analysis of variance were used to determine the statistical significance of the number of each bacterium and the detection rates between day 0, day 8 and day 15. Statistical significance was established at the P < 0.05 level.
Ethics.
This study was performed in accordance with the international guiding principles for biomedical research involving animals for the prevention of cruelty to animals and was approved by the ethical committee at our medical college.
|
RESULTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Regarding aerobic bacteria, Streptococcus significantly decreased just after treatment, while Enterococcus and Enterobacteriaceae significantly increased. However, the total number of aerobic bacteria was not significantly decreased on day 8. On day 15, the number of aerobic bacteria had increased, which might be due to an increase in bacteria that were resistant to the antimicrobials used in this study.
Regarding anaerobic bacteria, the number of Lactobacillus and the detection rate of Bifidobacterium, Peptostreptococcus and Veillonella significantly decreased during eradication treatment. In contrast, the number of Bacteroidaceae did not change, with the result that the total number of anaerobic bacteria was not statistically significantly changed, as Bacteroidaceae were in the majority among bacterial flora. Lactobacillus, Bifidobacterium, Peptostreptococcus and Veillonella had almost recovered to the pretreatment level by day 15. In this study, yeast, Pseudomonas aeruginosa and Clostridium difficile were not detected.
Physical status of the macaques during the experiment
The stool of an experimental macaque in the usual condition is solid and rather hard. All the macaques in the treatment group had soft stools during the experiment but no macaque had watery diarrhoea. Neither behavioural changes nor appetite loss was observed.
|
DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many studies have shown the influence of drugs, including antimicrobials, on the composition pattern of intestinal microflora (Brismar et al., 1991, 1993; Edlund et al., 1994, 2000a, b; Floor et al., 1994; Stark et al., 1996). Also, how eradication treatment for H. pylori infections affects intestinal microflora has been described (Adamsson et al., 1999; Bühling et al., 2001). However, because studies were performed on human subjects, methodological weaknesses could not be overcome, especially with respect to anaerobic techniques. In fact, reports on how antimicrobial treatment affects the profile of human anaerobic intestinal microflora do not necessarily reach agreement. For instance, amoxycillin is generally thought to have little effect on anaerobic flora (Brismar et al., 1993; Floor et al., 1994; Stark et al., 1996), while one report noted that only Lactobacillus and Bifidobacterium were suppressed by this antibiotic (Edlund et al., 1994). Further discordance is recognized regarding the effect of clarithromycin. Some reports cited that only a few anaerobic bacteria and not total anaerobic flora was suppressed during clarithromycin administration (Brismar et al., 1991; Edlund et al., 2000a), while another report showed that the total number of anaerobic bacteria decreased (Edlund et al., 2000b).
These discrepant observations in humans may partially come from methodological difficulties. With this background, we investigated changes in intestinal flora by antimicrobial treatment for H. pylori infections using Japanese macaques. Our study using this methodology has both merits and demerits.
One of the merits of this experimental animal is that its intestinal flora is known to be quite similar to that of human beings (Benno et al., 1987; Mitsuoka & Kaneuchi, 1977). In addition, we could analyse stool samples taken directly from the rectum, which enables use of an anaerobic technique. Furthermore, strict control of diet and environmental conditions in these experimental animals could minimize environmental bias.
On the other hand, using experimental animals might be a limitation, as the delivery, distribution and metabolization of the antimicrobials may differ from that in humans, and the amounts of the drugs given in our experiment may represent an overdose for macaques (body weight about 9 kg) since the amount was determined based on human use. Yet, higher doses of treatment drugs might be preferable since adverse effects of such treatment should be enhanced. In this experiment we used animals that were not infected with H. pylori. Regarding this concern, we previously reported that the intestinal microflora in H. pylori-infected macaques did not differ from those not infected (Fukuda et al., 1998).
Alterations in the composition of intestinal microflora were observed in this study during antimicrobial treatment; previous studies also made this observation. Adamsson et al. (1999) and Bühling et al. (2001) showed the change of human intestinal microflora following H. pylori treatment with omeprazole/metronidazole/amoxycillin and omeprazole/metronidazole/clarithromycin, respectively. Both demonstrated a marked decrease in anaerobic bacteria during treatment, but this decrease had almost reversed 35 days after the initiation of therapy. However, the eradication regimens used in both reports included metronidazole, which is not approved for H. pylori eradication therapy by the Japanese social security foundation. Metronidazole is known to possess potent suppressive effects, especially on anaerobic bacteria, and regimens that do not contain this antimicrobial may have a different effect on bacterial flora. Yet literature on regimens without metronidazole is not available. Therefore, we investigated the effect of the PPI/amoxycillin/clarithromycin regimen, which is one of the most popular regimens in the world (Sharma et al., 1999). In our study, the total number of anaerobic bacteria did not change by day 8, although specific species such as Lactobacillus and Bifidobacterium significantly decreased. These observations suggest that various combinations of antimicrobials yield different alterations in the composition of intestinal microflora.
Generally, diarrhoea and soft stools are considered to be results of disturbances of intestinal microflora, although the mechanism for diarrhoea is not fully understood (Bartlett, 2002). In clinical reports on diarrhoea and soft stools due to eradication treatment for H. pylori infection, the prevalence of these side effects differed according to the combination of antimicrobials. Lind et al. (1996) demonstrated in their MACH 1 (metronidazole, amoxicillin, clarithromycin, H. pylori, 1 week therapy) study that diarrhoea and soft stools appeared in 12.1–17.9 % of the patients treated with metronidazole-containing PPI-based regimens, whereas these side effects were seen in 26.6–30.0 % of those on regimens without metronidazole. Misiewicz et al. (1997) reported that the prevalences of diarrhoea and soft stool in two different regimens using metronidazole and one regimen without metronidazole were 4.2 %, 9.2 % and 18.2 %, respectively. These observations suggest that regimens without metronidazole may induce diarrhoea or soft stool more often than those with this drug. These findings, together with our results that anaerobic bacteria were not suppressed during treatment, suggest that whether anaerobic flora is suppressed may relate to presentation of diarrhoea or soft stools.
It should be noted that the alterations in bacterial microflora had almost reversed 7 days after completion of the treatment, suggesting that the clinically unfavourable effects such as diarrhoea or soft stools would stop in a short time. In fact, it is clinically well-known that the adverse effects of eradication treatment do not last for long periods (Armuzzi et al., 2001; Cremonini et al., 2002; Lind et al., 1996). Alterations in intestinal bacterial microflora in macaques by combination treatment for H. pylori infections supported this clinical experience.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
