The effect of antibiotic therapy on the gastrointestinal tract in children

Currently, increasing attention is being paid to the problem of alterations in gut microbiota, especially in children, under the influence of various environmental factors. Among these, the use of antibacterial drugs is the most obvious practical case.

The microbiota of the gastrointestinal tract (GI) exerts a significant influence on the health of the growing child's body, particularly in early childhood. The establishment and development of the gut microbiota begin at birth and continue throughout the first years of life. The formation of the GI microbiota is influenced by various factors, the leading ones being the mode of delivery (vaginal, cesarean section) and the infant's feeding characteristics (Fig. 1).

The duration of breastfeeding, the timing and nature of introduced complementary foods, the transition to mixed or formula feeding, and changes in habitual diet can lead to digestive disturbances in infants. With formula feeding, the infant's GI tract is more often colonized by bacteria of the Enterobacteriaceae spp., C. difficile, Bacteroides spp., and Streptococcus spp. families, unlike breastfed infants, whose intestines are predominantly colonized by Staphylococcus spp., Streptococcus spp., Lactobacillus spp., and Bifidobacterium spp. Most members of the GI microbiota belong to six dominant phyla: Firmicutes, Bacteroidetes, Proteobacteria, Verrucomicrobia, Actinobacteria, and Fusobacteria [1, 2].

It is known that antibacterial therapy negatively affects GI function in both children and adults. At the same time, among modern antibacterial agents, it is very difficult to find drugs that do not have an adverse effect on the digestive tract. In addition to gastric dyspepsia, manifested by nausea and vomiting, diarrhea most often develops in patients taking antibiotics. According to the literature, the frequency of adverse events with azithromycin use is about 9%, although for certain conditions it can reach up to 20% (e.g., in urethritis) [3, 4]. In most cases, azithromycin intake is associated with moderately pronounced GI reactions (abdominal pain, nausea, vomiting, diarrhea, which usually do not require drug discontinuation). Antibacterial therapy in young children promotes an increase in the proportion of enterobacteria and enterococci, which persists even one month after the end of therapy [5]. Antibiotic use during pregnancy or breastfeeding also affects the microbial diversity in infants [6]. Madan J.C. et al. (2012) demonstrated the potential for sepsis in newborns when the phylogenetic diversity of the gut microflora is reduced [7]. For example, a 5-day course of oral antibiotics alters the composition of the human gut microbiota for up to 4 weeks, after which the microbial balance is restored, returning to the original association characteristic of that ecotype; however, some bacterial associations in the gut do not recover within an observation period of up to 6 months [8].

In recent years, researchers have shown great interest not only in GI damage with the development of antibiotic-associated diarrhea but also in the link between changes in gut microbiota and the onset of diseases such as diabetes, obesity, allergies, etc. [9, 10]. It has been established that repeated courses of antibiotics in infants lead to a decrease in the number of Bifidobacteria and Bacteroides in the gut. Changes in the levels of these gut microbiota members are specifically linked to the risk of developing obesity. After a course of antibiotic treatment, the bifidobacteria population recovers slowly, while representatives of the Bacteroides spp. phylum usually do not fully recover [11].

Our department conducted an open comparative study in children with rhinosinusitis to evaluate the efficacy and safety of the drug Ecomed (powder for oral suspension, containing azithromycin 100 mg/5 ml, produced by JSC "AVVA RUS", Russia, in combination with lactulose) compared to conventional azithromycin without lactulose. The study included 100 patients (50 boys and 50 girls) aged 3 to 14 years, who were divided into 2 groups:

• The main group consisted of 50 children (mean age 6.16 ± 2.78 years) who received Ecomed for 3 days.
• The control group consisted of 50 children (mean age 6.9 ± 3.24 years) who received azithromycin for 3 days.

• Visit 1 (Screening): After signing the informed consent form, the patient was assigned an ascending screening identification number. Medical history and complaints were collected, followed by a physical examination and anthropometric measurements of the child. A separate assessment of the functional state of the gastrointestinal tract was performed, considering the following symptoms: abdominal pain, bloating and flatulence, nausea, dyspepsia, frequency of bowel movements, and stool consistency.
• Visit 2 (End of Therapy): This visit was conducted after the completion of antibacterial therapy. It involved collecting current complaints, performing a physical examination, measuring body temperature, assessing the functional state of the GI tract, and recording any concomitant therapy.
• Visit 3 (Observation Period): This visit took place two weeks after the end of antibacterial therapy. It included collecting complaints, performing a physical examination of the patient, measuring body temperature, and assessing symptoms related to the functional state of the GI tract.
• Visit 4 (Observation Period): This visit was conducted two months after the completion of the antibacterial therapy course.

During each visit, a stool sample was collected for analysis – specifically, for 16s rRNA gene sequencing. DNA was extracted from 180–220 mg of frozen fecal samples using the commercial QIAmp DNA Stool Mini Kit (Qiagen, Germany). For samples that passed quality control, a library was prepared following the Metagenomic Sequencing Library Preparation Manual (Illumina). Library sequencing was performed on the MiSeq platform (Illumina).

Picture 1

The functional state of the gastrointestinal tract was assessed before (Visit 1) and after antibacterial therapy (Visit 2), as well as during follow-up (2 weeks and 2 months after the end of antibacterial therapy - Visits 3 and 4) using a scoring system based on the following criteria: abdominal syndrome, flatulence, bloating, changes in stool character, and dyspeptic symptoms. To assess the severity of these disturbances, a scoring system was used where 0 indicates the absence of a symptom, 1 indicates a mild symptom, 2 indicates a moderate symptom, and 3 indicates a severe symptom.

Stool consistency was assessed according to the following criteria: 0 – liquid, 1 – mushy, 2 – semi-formed, 3 – formed. To assess the frequency of bowel movements, the following gradation was used: 0 – up to 3 times, 1 – more than 4 times per day, 2 – more than 5 times per day, 3 – more than 7 times per day.

In the assessment of functional GI disorders in patients before antibacterial therapy, no statistically significant differences were observed between the groups for the criteria of abdominal pain (p = 0.157), bloating (p = 1.0), flatulence (p = 1.0), and stool consistency (p = 0.458). Thus, at the time of inclusion in the study, there were no statistically significant differences between the two study groups.

We assessed the severity and frequency of functional gastrointestinal disorders after the completed course of antibacterial therapy in the groups (Fig. 2).

After the course of antibacterial therapy, assessment of functional GI disorders revealed a statistically significant difference between the groups for the following indicators: flatulence was present in 42% of children in the azithromycin group compared to 14.3% in the Ecomed group (p = 0.002), stool consistency (p = 1.016E-06), and stool frequency more than 3 times per day was recorded in 28% of children in the azithromycin group compared to 0% in the Ecomed group (p = 0.000042).

At the third visit (approximately 14 ± 3 days after the end of therapy), statistically significant differences were observed only for stool consistency – deviations in 38% vs. 16.3% (p = 0.007).

In the main patient group, two months after the end of the antibacterial therapy course, mild abdominal pain was reported by 3 children (6%), bloating by 9 patients (18%), and flatulence by 4 children (8%). Abnormal stool character in the form of liquid stool was recorded in only 1 child (2%), semi-formed in 3 children (6%), and mushy in 2 children (4%). It is important to note that increased stool frequency was not observed in any child in the main group throughout the entire observation period.

In the control patient group at the 4th visit, mild abdominal pain persisted in 4 children (8%), bloating in 5 patients (10%), and flatulence in 8 children (16%). Moderately pronounced bloating was noted in 2 children (4%) and flatulence in one patient. Complaints of liquid stool were made by only one child, mushy stool by 2 children (4%), and semi-formed stool persisted in 9 children (18%) in the control group (Figs. 2, 3, 4). Thus, two months after the end of antibacterial therapy, no statistically significant differences between the groups for the studied indicators could be found.

• For the indicator "bloating," significant differences in both groups were recorded from Visit 1 to Visit 2 (p = 0.001 for the azithromycin group, p = 0.049 for the Ecomed group), and from Visit 1 to Visit 3 (p = 0.008 for the azithromycin group, p = 0.039 for the Ecomed group). However, in the Ecomed group, a more "smoothed" course of "bloating" was observed, since, unlike the azithromycin group, no statistically significant differences were found from Visit 3 to Visit 2, from Visit 4 to Visit 2, and from Visit 4 to Visit 3.
• For the indicator "stool consistency," significant differences were obtained only in the azithromycin group – from Visit 1 to Visit 3 (p = 0.003), from Visit 3 to Visit 2 (p = 0.000), from Visit 4 to Visit 2 (p = 0.000), and from Visit 4 to Visit 3 (p = 0.004). In the Ecomed group, changes in stool consistency were not significant.
• In the Ecomed group, changes in the "defecation frequency" indicator were not significant between visits, while in the azithromycin group, significant differences were obtained – from Visit 1 to Visit 2 (p = 0.000), from Visit 3 to Visit 2 (p = 0.004), and from Visit 4 to Visit 2 (p = 0.001).

Analysis of gut microflora biodiversity by visits within groups showed no statistically significant differences between visits at the family, genus, and species levels. Nevertheless, the mean values and medians in the Ecomed group for genera and families tended towards greater diversity at all visits except for species.

To determine statistically significant changes in the bacterial composition of the microbiota, microorganisms with an average representation in the samples of at least 0.1% on at least one visit were selected. Furthermore, microorganisms with a frequency of occurrence in the group for any visit equal to or exceeding 30% were selected.

Based on the subsequent comparison of the abundance data of microorganisms selected by the method described above between visits, as well as when comparing the magnitude of abundance changes considering their direction in the two groups, the following significant genera were identified: Ruminococcus, Prevotella, Enterobacter, Roseburia, Streptococcus, Alkaliphilus, Parabacteroides, Lachnospira, Enterococcus, Sutterella, Veillonella, Odoribacter, Lactobacillus, Paraprevotella, Anaerostipes, Dysgonomonas (Fig. 3), as well as the species: Prevotella copri, Blautia coccoides, Ruminococcus bromii, Alkaliphilus crotonatoxidans, Streptococcus vestibularis, Ruminococcus gnavus, Lachnospira pectinoschiza, Blautia wexlerae, Streptococcus thermophilus, Coprococcus eutactus, Bacteroides sartorii, Blautia hansenii, Bifidobacterium bifidum, Enterococcus lactis, Eggerthella lenta, Erysipelothrix inopinata, Streptococcus parasanguinis, Clostridium frigoris, Adlercreutzia equolifaciens, Negativicoccus succinicivorans, Bacteroides paurosaccharolyticus, Streptococcus bovis/gallolyticus, Ruminococcus albus (Fig. 4). Analysis of the total abundance showed that the drugs affect 20–25% of the total gut microflora (based on fecal composition analysis).

Analysis of changes in microbial abundance revealed distinct microbiota profiles during therapy with azithromycin and Ecomed. The observed effects were divided into several qualitative categories: inhibition, stimulation, and preservation of microbiota.

Inhibitory and Stimulatory Effects of Azithromycin. The latter statistically significantly inhibited the abundance of the genera Enterobacter, Sutterella, Odoribacter, Paraprevotella, Parabacteroides, as well as the species Coprococcus eutactus, Eggerthella lenta, Bifidobacterium bifidum, Bacteroides paurosaccharolyticus, and Negativicoccus succinicivorans. At the same time, an increase in the abundance of Ruminococcus gnavus, Ruminococcus albus, the genus Streptococcus and its species Streptococcus vestibularis, Streptococcus bovis, Streptococcus thermophilus, Streptococcus parasanguinis, as well as Alkaliphilus and its species Alkaliphilus crotonatoxidans, Lachnospira and its species Lachnospira pectinoschiza, Veillonella, Blautia wexlerae, and Adlercreutzia equolifaciens was recorded.

Azithromycin inhibits the abundance (p < 0.05) of the genus Prevotella and its species Prevotella copri, but stimulates the abundance (p < 0.05) of Ruminococcus, its species Ruminococcus bromii, the genera Roseburia, Enterococcus and its species Enterococcus lactis and Enterococcus durans, Lactobacillus, Anaerostipes, Parabacteroides and the species Blautia coccoides, Blautia hansenii, Bacteroides sartorii, Bifidobacterium bifidum, Eggerthella lenta, and Erysipelothrix inopinata.

Based on a comparison of the inhibitory and stimulatory effects of azithromycin and lactulose, microorganisms that were preserved during Ecomed therapy, unlike with azithromycin alone, were identified. Therapy with Ecomed prevented a significant increase in the population of the genus Streptococcus, as well as the species Streptococcus vestibularis, Streptococcus thermophilus, Streptococcus bovis, Streptococcus parasanguinis, Bacteroides sartorii, Bacteroides paurosaccharolyticus, Adlercreutzia equolifaciens, and Ruminococcus gnavus. Furthermore, unlike azithromycin, Ecomed did not cause a decrease in the abundance of the genera Sutterella, Odoribacter, Paraprevotella, or the species Coprococcus eutactus and Negativicoccus succinicivorans.

In addition to these observations, it can be noted that lactulose does not have a protective effect on Enterobacter – by the later visits, a significant inhibition of their abundance occurred in both drug groups. In the azithromycin group, suppression of the abundance of the genera Enterobacter, Sutterella, and the species Bifidobacterium bifidum (and Bacteroides in general) was recorded, which is consistent with data on the effect of azithromycin on these taxa [12–14].

The abundance of Enterobacter decreased more than threefold in both groups. This genus belongs to the commensal microflora of the human GI tract and is generally considered pathogenic only in patients with reduced resistance to infections or immune system disorders [15, 16]. Since a decrease in the mean abundance of Enterobacter is also observed in the Ecomed group, this effect is most likely realized through the direct antibacterial action of azithromycin. It is noteworthy that lactulose as part of Ecomed did not significantly influence the preservation of the mean abundance of the genus and species of Enterobacter. Most likely, lactulose is not a primary nutritional substrate for bacteria of the genus Enterobacter, which is somewhat confirmed by literature data [17].

In the azithromycin group, suppression of Odoribacter was recorded. These are anaerobic bacteria that, in addition to the lower GI tract, colonize the oral mucosa, can cause bad breath (hence the name), and are also a cause of abscesses and periodontitis in humans. Also, the mean abundance of Paraprevotella decreased in the group of patients receiving azithromycin, unlike the Ecomed group, by visit 4. This genus is one of the most dominant genera in the GI tract, while exhibiting properties of opportunistic pathogens. Their main metabolites are succinate and acetate, and they utilize various sugars. It is known that increased levels of this microorganism are associated with a diet rich in starch, sugary cereals, and vegetables, and with a vegetarian diet [18]. In our study, the preservation of Paraprevotella in the Ecomed group may be related to the stimulatory effect of lactulose in Ecomed.

In the Ecomed group, stimulation of Ruminococcus growth occurs, likely due to the species Ruminococcus bromii. Ruminococcus utilize cellulose, produce acetic and lactic acids; furthermore, they ferment lactulose [21, 22].

Unlike the azithromycin group, an increase in the abundance of the genus Lactobacillus was detected in the Ecomed group in stool samples collected by visit 3. Probiotic strains of the genus Lactobacillus are normal flora of the human GI tract; they reduce the level of intestinal gases; exhibit anti-inflammatory action; reduce the amount of toxins; stimulate intestinal motility, and have the ability to ferment lactulose [23]. Their main metabolites are lactic acid and bacteriocins, which mediate the antibacterial properties of this genus. In our study, despite information in the literature about the antibiotic susceptibility of these microorganisms, no significant decrease in the mean abundance of Lactobacillus was observed, possibly due to their initially low abundance in our study population, as well as the resistance of these microorganisms to azithromycin.

Unlike the azithromycin group, an increase in the abundance of the genus Lactobacillus was detected in the Ecomed group in stool samples collected by visit 3. Probiotic strains of the genus Lactobacillus are normal flora of the human GI tract; they reduce the level of intestinal gases; exhibit anti-inflammatory action; reduce the amount of toxins; stimulate intestinal motility, and have the ability to ferment lactulose [23]. Their main metabolites are lactic acid and bacteriocins, which mediate the antibacterial properties of this genus. In our study, despite information in the literature about the antibiotic susceptibility of these microorganisms, no significant decrease in the mean abundance of Lactobacillus was observed, possibly due to their initially low abundance in our study population, as well as the resistance of these microorganisms to azithromycin.

The suppression of opportunistic pathogenic flora growth in the Ecomed group is of particular interest. The main mechanism for inhibiting streptococcal growth is mediated through the activity of other microorganisms. This phenomenon can be realized through different pathways. First, through competition for substrate. Although streptococci are capable of fermenting lactulose and other substances to a small extent, other bacteria contain specific enzymes that allow for more active utilization (e.g., alpha- and beta-galactosidases). These include lactobacilli, clostridia [25], Anaerostipes and Blautia [26]. Indeed, in the Ecomed group, we observe a statistically significant increase, particularly in Lactobacillus (Visit 3, p = 0.03), due to the presence of the prebiotic component, unlike the azithromycin group. Second, there is fairly convincing data that lactobacilli can directly inhibit the growth of streptococci, thanks to the production of metabolites such as lactic acid, bacteriocins, etc. [27–29].

It is important to note that a reduced ability to ferment complex carbohydrates (by the remaining part of the microbial community in the azithromycin group) is a predisposing factor for diarrhea [32]. Consequently, patients in the azithromycin group become more vulnerable to opportunistic bacterial infections due to poor microbiota recovery compared to the lactulose group. Lactulose as part of the drug Ecomed maintains the population level of complex carbohydrate-fermenting microorganisms that produce short-chain fatty acids (SCFAs), which play an important role in maintaining gut homeostasis [33], and also help prevent the development of osmotic diarrhea and other GI disorders, including the overall degree of dysbiosis. Specifically, it prevents the decline in the abundance of the genera Sutterella, Odoribacter, Paraprevotella, and the species Coprococcus eutactus and Negativicoccus succinicivorans. Furthermore, lactulose in Ecomed helps restrain the growth of opportunistic pathogenic microflora.

• Unlike the Ecomed group, the azithromycin group showed a significant increase in the abundance of opportunistic microorganisms from the genus Streptococcus (p=0.04) and the species Streptococcus vestibularis (p = 0.03), Streptococcus bovis/gallolyticus (p = 0.0005) two months after the end of antibiotic therapy. Additionally, the azithromycin group exhibited a significant decrease in the abundance of beneficial microflora, namely the genus Enterobacter, as well as the genera Sutterella, Odoribacter, Paraprevotella, and the species Coprococcus eutactus, Bacteroides paurosaccharolyticus, Negativicoccus succinicivorans, which accounted for up to 5% of the total microbial abundance.
• The Ecomed group demonstrated prevention of opportunistic pathogenic flora growth and preservation of the abundance of beneficial and other microorganisms. Lactulose inhibits the growth of opportunistic flora through competitive displacement by utilizing microorganisms like Lactobacillus, thereby restoring the population of complex carbohydrate-fermenting microorganisms with anti-inflammatory properties.

1. Khan I. Implication of Gut Microbiota in Human Health. CNS & neurological disorders drug tar- gets, 2014, 13(8): 1325–33.

2. Ubeda C et al. Vancomycin-Resistant Enterococcus Domination of Intestinal Microbiota Is Enabled by Antibiotic Treatment in Mice and Precedes Bloodstream Invasion in Humans. Journal of Clinical Investigation, 2010, 120(12): 4332–41.

3. Kacmar J, Cheh E, Montagno A, and Peipert JF. A Randomized Trial of Azithromycin versus Amoxicillin for the Treatment of Chlamydia Trachomatis in Pregnancy. Infectious diseases in obstetrics and gynecology, 2001, 9(4): 197–202.

4. Takahashi, Satoshi et al. Clinical Efficacy of a Single Two Gram Dose of Azithromycin Extended Release for Male Patients with Urethritis. Antibiotics (Basel, Switzerland), 2014, 3(2): 109–20.

5. Tanaka S. et al. Influence of antibiotic exposure in the early postnatal period on the development of intestinal microbiota. FEMS Immunol. Med. Microbiol., 2009, 56: 80–87.

6. Fallani M et al. Intestinal microbiota of 6-week- old infants across Europe: geographic influence beyond delivery mode, breast-feeding, and anti- biotics. J. Pediatr. Gastroenterol. Nutr. 2010, 51: 77–84.

7. Madan JC et al. Gut microbial colonisation in premature neonates predicts neonatal sepsis. Arch. Dis. Child. Fetal Neonatal Ed, 2012, 97: F456–F462.

8. Dethlefsen L, Huse S, Sogin ML, Relman DA. The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16S rRNA sequencing. PLoS Biol, 2008, 6: 2383–2400.

9. Bulloch MN and Carroll DG. When One Drug Affects 2 Patients. Journal of Pharmacy Practice, 2012, 25(3): 352–67.

10. Ubeda C and Pamer EG. Antibiotics, Microbiota, and Immune Defense. Trends in Immunology, 2012, 33(9):459–66.

11. Penders J, Thijs C, Vink C, Stelma FF, Snijders B, Kummeling I, van den Brandt PA, Stobberingh EE. Factors influencing the composition of the intestinal microbiota in early infancy. Pediatrics, 2006, 118: 511–521

12. Brenner DJ, Krieg NR. Bergey’s Manual® of Systematic Bacteriology, Second Edition: Volume Two: The Proteobacteria, Part C. Springer Science & Business Media. 2006.

13. Wexler HM, Molitoris E, Molitoris D, and Finegold SM. In Vitro Activity of HMR 3004 (RU 64004) against 502 Strains of Anaerobic Bacteria. Anaerobe, 1999, 5(2): 65–68.

14. Stock I and Wiedemann B. Natural Antibiotic Susceptibility of Enterobacter Amnigenus, Enterobacter Cancerogenus, Enterobacter Gergoviae and Enterobacter Sakazakii Strains. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases, 2002, 8(9): 564–78.

15. Barnes BJ, Wiederhold NP, Micek ST, Polish LB, and Ritchie DJ. Enterobacter Cloacae Ventriculitis Successfully Treated with Cefepime and Gentamicin: Case Report and Review of the Literature. Pharmacotherapy, 2003, 23(4): 537–42.

16. Lupp C et al. Host-Mediated Inflammation Disrupts the Intestinal Microbiota and Promotes the Overgrowth of Enterobacteriaceae. Cell host & microbe, 2007, 2(3): 204.

17. Oberhofer TR and Hajkowski R. Evaluation of Non-Lactose-Fermenting Members of the Klebsiella-Enterobacter-Serratia Division. II. Antibiotic Susceptibility. American journal of clinical pathology, 1970, 54(5): 726–32.

18. Rani V, Gagan D, Singh RK, Palle K, and Yadav UCS. Oxidative Stress and Metabolic Disorders: Pathogenesis and Therapeutic Strategies. Life sciences, 2016, 148: 183–93.

19. Băncescu G, Băncescu AA, Didilescu AC, & Constantinescu MV. Antimicrobial Susceptibility of Prevotella Isolates from Abscesses of Fascial Spaces of the Face and Neck. Revista Română de Medicină de Laborator, 2009, 17(4).

20. Boyanova L, Kolarov R, Gergova G, Dimitrova L, and Mitov I. Trends in Antibiotic Resistance in Prevotella Species from Patients of the University Hospital of Maxillofacial Surgery, Sofia, Bulgaria, in 2003–2009. Anaerobe, 2010, 16(5): 489–92.

21. Ардатская М.Д., Минушкин О.Н. Современные Принципы Диагностики И Фармакологической Коррекции. Гастроэнтерология, приложение к журналу Consilium Medicum, 2006, 8(2).

22. Moore WE and Moore LH. Intestinal Floras of Populations That Have a High Risk of Colon Cancer. Applied and environmental microbiology, 1995, 61(9): 3202–7.

23. Guerra-Ordaz AA et al. Lactulose and Lactobacillus Plantarum, a Potential Complementary Synbiotic to Control Postweaning Colibacillosis in Piglets. Applied and environmental microbiology, 2014, 80(16): 4879–86.

24. Korpela K, SalonenA, Virta LJ et al. Intestinal Microbiome Is Related to Lifetime Antibiotic Use in Finnish Pre-School Children. Nature communications, 2016, 7: 10410.

25. Hafez MM. Interference between Lactobacilli and Group A Streptococcus Pyogenes: An Expansion to the Concept of Probiotics. ew Egyptian Journal of Microbiology, 2007, 17(1): 262–84.

26. Kõll P. Characterization of Oral Lactobacilli as Potential Probiotics for Oral Health. Oral microbiology and immunology, 2008, 23(2): 139–47

27. Simark-Mattsson C, Jonsson R, Emilson C-G, and Roos K. Final pH Affects the Interference Capacity of Naturally Occurring Oral Lactobacillus Strains against Mutans Streptococci. Archives of oral biology, 2009, 54(6): 602–7.

28. Van den Bogert B et al. Comparative Genomics Analysis of Streptococcus Isolates from the Human Small Intestine Reveals Their Adaptation to a Highly Dynamic Ecosystem. PloS one, 2013, 8(12): e83418.

29. Zoetendal EG, Raes J, van den Bogert B, Arumugam M, Booijink CCGM et al. The Human Small Intestinal Microbiota Is Driven by Rapid Uptake and Conversion of Simple Carbohydrates. The ISME journal, 2012, 6(7): 1415–26.

30. Young VB and Schmidt TM. Antibiotic-Associated Diarrhea Accompanied by Large-Scale Alterations in the Composition of the Fecal Microbiota. Journal of clinical microbiology, 2004, 42(3): 1203–6.

31. Pryde SE, Duncan SH, Hold GL, Stewart CS, and Flint HJ. The Microbiology of Butyrate Formation in the Human Colon. FEMS microbiology letters, 2002, 217(2): 133–39.