Different studies have investigated changes in microbiota composition and their implications in different IBS subtypes (Table 12).

Author (Year)	Country	Study Type	Sample Size	IBS Subtype	Type of Analysis	Sample Type	Rome Criteria	Results
Carroll et al. (2011) [1][77]	USA	Prospective monocentric	37 (16/21)	IBS-D	16S rRNA T-RFLP	Fecal Mucosal	Rome III	Higher level of microbial biodiversity in fecal- than in mucosal-associated communities within IBS-D (p = 0.008)
Rajilić-Stojanović et al (2011) [2][87]	Europe	Prospective monocentric	108 (62/46)	IBS-D IBS-C IBS-M	16S rRNA qPCR	Fecal	Rome II	IBS-C had increased Firmicutes (including Clostridium spp.) (p < 0.05) and decreased Actinobacteria and Bacteroidetes (p < 0.01) vs. controls
Durbàn et al. (2012) [3][85]	Europe	Prospective monocentric	16/9	IBS-D IBS-C	16S rRNA qPCR	Fecal Mucosal	Rome II	IBS-D had increased fecal Acinetobacter (OR = 16.7, p = 0.02), Butyricimonas (OR = 2.29, p = 0.004), and Odoribacter (OR = 6.11, p = 0.003), but decreased mucosal Oribacterium (OR = 0.17, p = 0.04), Brevundimonas (OR = 0.09, p = 0.0009), and Butyricicoccus (OR = 0.38, p = 0.026) vs. controls. IBS-C had increased fecal Alistipes (OR = 5.82, p = 0.01) and Butyricimonas (OR = 3.27, p = 0.004) and increased mucosal Bacteroides (OR = 3.15, p = 0.003), but decreased Coprococcus (OR = 0.03, p = 0.007), Fusobacterium (OR = 0.02, p = 0.003), Streptococcus (OR = 0.06, p = 0.007), and Veillonella (OR = 0.03, p = 0.04) in feces vs. controls.
Carroll et al. (2012) [4][80]	USA	Prospective monocentric	46 23/23	IBS-D	16S rRNA qPCR	Fecal	Rome III	Higher Enterobacteriaceae (p = 0.03) and lower Fecalibacterium genera (p = 0.04)
Parkes et al. (2012) [5][88]	Europe	Prospective monocentric	47/26	IBS-D IBS-C	FISH	Mucosal (rectum)	Rome III	IBS-D had lower bifidobacteria vs. IBS-C and controls (p = 0.011). In IBS, maximum stools/day negatively correlated with mucosal Bifidobacteria (p < 0.001) and Lactobacilli (p = 0.002).
Chassard et al. (2012) [6][89]	Europe	Prospective multicentric	14/12	IBS-C	16S rRNA FISH	Fecal	Rome II	IBS-C had lower lactate-producing/utilizing bacteria and methanogens/acetogens (p < 0.05), but 10–100× higher H2/lactate-utilizing sulfate reducers vs. controls. Roseburia-E. rectale butyrate producers were lower in IBS-C (p < 0.05–0.01). Mucosal vs. fecal microbiota differed significantly (p = 0.002) regardless of IBS characteristics. Mucosal microbiota was dominated by Bacteroidetes, fecal by Firmicutes/Actinobacteria/Proteobacteria. Controls had higher uncultured Clostridiales (p < 0.005) in mucosa than IBS.
Rangel et al. (2015) [7][78]	Europe	Retrospective monocentric	49 33/16	IBS-D IBS-C IBS-M IBS-U	16S rRNA Phylogenetic microarray	Fecal Mucosal	Rome III	Mucosal vs. fecal microbiota differed significantly (p = 0.002), independent of IBS characteristics. Mucosa was dominated by Bacteroidetes, feces by Firmicutes/Actinobacteria/Proteobacteria. Controls had higher uncultured Clostridiales (p < 0.005) in mucosa vs. IBS. Fecal bacterial diversity higher than mucosal in IBS (p < 0.005).
Pozuelo et al. (2015) [8][86]	Europe	Prospective multicentric	113/66	IBS-D IBS-C IBS-M	16S rRNA qPCR	Fecal	Rome III	IBS had lower microbial diversity associated with lower butyrate-producing bacteria, especially in IBS-D/M (p = 0.002). Untreated IBS had lower Methanobacteria vs. controls (p = 0.005). Bacterial taxa was correlated with flatulence/abdominal pain (p < 0.05).
Shukla et al. (2015) [9][83]	Asia	Prospective monocentric	47/30	IBS-D IBS-C IBS-U	16S rRNA qPCR	Fecal	Rome III	Relative difference of Bifidobacterium (p = 0.042) was lower, while Ruminococcus productus-Clostridium coccoides (p = 0.016), Veillonella (p = 0.008), Bacteroides thetaiotamicron (p < 0.001), and Pseudomonas aeruginosa (p < 0.001) were higher among IBS patients than controls. Lactobacillus (p = 0.002) was lower, while Bacteroides thetaiotamicron (p < 0.001) and segmented filamentous bacteria (SFB, p < 0.001) were higher among IBS-D than IBS-C. Numbers of Bacteroides thetaiotamicron (p < 0.001), P. aeruginosa (p < 0.001), and Gram- (p < 0.01) were higher among IBS-C and IBS-D than controls. Quantity of SFB was higher among IBS-D (p = 0.011) and lower among IBS-C (p = 0.002). Veillonella species was higher among IBS-C than controls (p = 0.002).
Liu et al. (2017) [10][84]	Europe Asia USA	Systematic review and meta-analysis	360 (13 studies)	IBS-D IBS-C IBS-M	qPCR	Fecal Mucosal	Rome II Rome III	Subgroup analysis showed IBS-D patients had significantly different expression of Lactobacillus (SMD = −1.81, p < 0.001) and Bifidobacterium (SMD = −1.45, p < 0.001).
Zhuang et al. (2018) [11][79]	Asia	Prospective monocentric	43 30/13		16S rRNA pyrosequencing	Fecal	Rome III	IBS-D had decreased fecal microbiota richness (p < 0.05) but not diversity vs. controls: Bacteroidetes (64.6%), Firmicutes (26.1%), Fusobacteria (5.2%), and Proteobacteria (3.7%).
Li et al. (2018) [12][81]	Asia	Prospective Monocentric	33/15	IBS-D	16S rRNA pyrosequencing	Mucosal (duodenum + rectum)	Rome III	Mucosal microbiota in duodenal samples differed from rectal samples in HC (p = 0.003), while less difference was shown in IBS-D (p = 0.052). Identified 24 genera were shared in duodenum and rectum, which both changed in IBS-D.
Sun et al. (2019) [13][102]	Europe, USA, Asia, Australia	Systematic review and meta-analysis	448 (15 studies)	IBS-D IBS-C IBS-M IBS-U	16S rRNA	Fecal	Rome I Rome II Rome III	IBS had higher fecal SCFAs vs. controls (SMD = 0.44). IBS-C had lower propionate (SMD = −0.91) and butyrate (SMD = −0.53) than controls. IBS-D had higher butyrate than controls.
Wang et al. (2020) [14][41]	Europe USA Asia	Systematic review and meta-analysis	208	IBS-D	16S rRNA qPCR	Fecal	Rome IV	Lower Lactobacillus (MD = −0.62 log10CFU/g); Lower Bifidobacterium (MD = −0.86 log10CFU/g); Higher E. coli (MD = −40.77 log10CFU/g); Lower Lactobacillus (MD = −0.43 log10CFU/g); Lower Bifidobacterium (MD = −1.76 log10CFU/g).
			105	IBS-C		Fecal

Specifically, one study with 16 patients and 21 controls reported a significant 1.2-fold lower diversity (p = 0.008) in IBS-D patients compared to healthy controls ^[15][76]. Regarding the microbial diversity, according to one study on fecal and mucosal samples from 33 IBS patients, mostly with IBS-D subtype (17/33), and 16 healthy controls, bacterial diversity was higher in fecal samples compared with mucosal ones (corrected p < 0.005) ^[1][77].

In contrast, another study reported a lower microbiota richness in the fecal samples of 27 IBS-D patients (p < 0.05), but no significantly lower diversity between these cases and healthy controls. Moreover, the major represented phyla in IBS-D microbiota were Bacteroidetes (64.6%), Firmicutes (26.1%), Fusobacteria (5.2%), and Proteobacteria (3.7%), albeit Bacteroidetes (56.4%), Firmicutes (35.9%), Proteobacteria (5.6%), and Fusobacteria (1.4%) were higher in the healthy subjects’ samples ^[7][78].

In terms of bacterial variety, the analysis of fecal microbiota of 23 IBS-D patients and 23 healthy controls resulted in significantly superior levels of the Enterobacteriaceae family (p = 0.03), and minor levels of the Faecalibacterium genera (p = 0.04) in the patient group compared to healthy subjects ^[11][79]. Moreover, one study conducted on 33 IBS-D patients and 15 healthy subjects analyzed the microbiota composition in different sites, such as duodenal and rectal mucosal samples. A significant difference between the duodenal and rectal microbiota in healthy subjects was reported (p = 0.003), while this difference was much lower in the IBS-D patients’ group (p = 0.052) ^[4][80]. Analyzing the fecal and small intestine mucosal microbiota of 37 IBS patients compared to 20 healthy subjects, P. aeruginosa was more frequently detected both in fecal samples (2.34% of total bacterial load) and small bowel samples (8.3% of total bacterial load) of patients with IBS compared to healthy subjects (p < 0.001) ^[12][81]. Another study involving 47 patients with IBS (20 with IBS-D, 20 with IBS-C, and 7 with IBS-U) analyzed fecal samples with quantitative polymerase chain reaction (qPCR), showing a significantly lower number of Lactobacillus in IBS-D patients than IBS-C patients (p = 0.002), while Bacteroides thetaiotamicron and segmented filamentous bacteria (from Bacillota phylum) were more numerous (corrected p = 0.001). Moreover, the species P. aeruginosa was more frequent in both IBS-D and IBS-C patients (97.9%), compared to healthy controls (33.3%; corrected p = 0.001) ^[16][82].

A systematic review and meta-analysis of 13 studies including 360 IBS patients and 268 healthy controls, only considering the qPCR analysis, compared the different IBS subtypes in terms of the microbiota. According to the presented data, a significant difference in IBS patients was reported for Lactobacillus, Bifidobacterium, and Fecalibacterium prausnitzii (p < 0.001), but not for Bacteroides-Prevotella, Enterococcus, E. coli, and C. coccoides, and this difference was especially driven by the IBS-D after subgroup analysis ^[9][83].

When comparing mucosal and fecal microbiota in IBS patient subgroups (13 IBS-D, 3 IBS-C, and 9 healthy subjects), a higher number of members of the Enterobacteriaceae and Rikenellaceae family were reported, respectively, in mucosal and fecal samples of IBS-C patients; more specifically, an increase of Alistipes and Butyricimonas were found in the first ones, while, in second ones, higher Bacteroides and lower Coprococcus, Eubacterium, Fusobacterium, Haemophilus, Neisseria, Odoribacter, Streptococcus, and Veillonella counts were registered. Instead, in the IBS-D subtype, there was an over-representation of Acinetobacter, Butyricimonas, Leuconostoc, and Odoribacter in feces and a reduction in Desulfovibrio, Oribacterium, Brevundimonas, and Butyricicoccus in the mucosal colonic samples ^[10][84]. In a larger study on 113 patients’ and 66 controls’ fecal samples, significantly lower butyrate-producing bacteria, such as Ruminococcaceae, unknown Clostridiales, Erysipelotrichaceae, and Methanobacteriaceae, were observed in IBS-D and IBS-M patients, when compared to the healthy group (corrected p = 0.002), while the microbiome of IBS-C patients did not show significant differences ^[3][85].

Concerning, more specifically, the IBS-C subtype, another research work analyzed through qPCR fecal samples of 62 patients with the IBS-D, IBS-C, and IBS-M subtypes compared to 46 healthy controls. The study reported significantly higher levels of Firmicutes, such as Clostridium species (p < 0.05), in IBS-C patients, and significantly reduced levels of Actinobacteria and Bacteroidetes phyla (p < 0.01) in this disease subtype ^[8][86].

Moreover, the analysis of rectal mucosa-associated microbiota of 27 IBS-D and 20 IBS-C patients compared to 26 healthy subjects showed higher levels of Bacteroidetes, Bifidobacterium, and C. coccoides/E. rectale (corrected p = 0.003), while bifidobacteria were lower in the IBS-D (corrected p = 0.011) group than in the IBS-C group and healthy controls ^[2][87]. From the analysis of the fecal microbiota of 14 IBS-C patients and 12 healthy subjects, several bacterial populations significantly differed between IBS-C patients and healthy controls ^[5][88]. The numbers of lactate-producing/utilizing bacteria and hydrogen-consuming microbes were at least 10× lower in IBS-C (p < 0.05). Conversely, sulphate-reducing bacteria utilizing lactate/hydrogen were 10–100× higher. The butyrate producer Roseburia-Eubacterium rectale was also lower in IBS-C (0.01 < p < 0.05). Fecal samples from IBS-C produced more sulphides and hydrogen, and less butyrate during starch fermentation than controls ^[5][88].

IBS-associated microbiota was also compared to some organic diseases. For instance, a research work evaluated IBS-D patients compared to ulcerative colitis (UC) patients. According to this trial involving 20 IBS-D patients, 28 UC patients (16 active and 12 inactive), and 16 healthy subjects, after a count of mucosa-associated microbiota using the fluorescent in situ hybridization (FISH) of mucosal biopsies, E. coli, Clostridium, and Bacteroides were significantly higher in IBS-D and UC patients, while Bifidobacteria were lower in UC and IBS patients compared to controls (p < 0.05) ^[6][89].

Other studies focused on the comparison between IBS, particularly IBS-D, and inflammatory bowel diseases ^[17][18][19][90,91,92]. Notably, in a model of human-microbiota-associated rats (HMAR) with induced experimental colitis, the IBS-C bacterial signature has shown anti-inflammatory properties with a reduction of pro-inflammatory cytokines ^[20][93].

Moreover, biofilms, a recently investigated entity recognized as an endoscopic finding in IBS and IBD patients, are considered a possible contributing factor in the IBS pathophysiology ^[21][94]. According to a study on 56 patients with IBS, 25 with IBD (specifically UC) compared to 36 healthy controls, biofilms are associated with a significant ten-fold elevation of primary bile acids in the intestinal lumen of IBS patients ^[22][95].

Even non-intestinal diseases were hypothesized to show similarities to IBS microbiota, specifically in a study which evaluated the microbiota changes in patients with IBS-D comparing to patients with depression by the analysis of fecal samples of 80 patients and 20 controls, reporting similar alterations in patients with IBS-D and depression, like higher Bacteroidetes and lower Firmicutes phylum levels ^[23][96].

In the context of IBS subtypes and clinical manifestations, the correlation between gut microbiota and symptoms severity was also investigated in several studies. In a research work on 110 IBS patients and 39 healthy controls, severe IBS patients exhibited lower microbial richness, a lower count of Prevotella enterotype and Methanobacteriale, and an increase in Bacteroides ^[24][36]. In another study with 80 IBS patients and 20 controls, IBS patients complaining of bloating had higher levels of Ruminococcaceae compared to patients without bloating (p < 0.05) ^[25][97]. Abdominal pain was also correlated with certain microbial taxa. Positive correlations were found with Bacteroides (p =  0.002), Ruminococcus (p = 0.004), and an unknown Barnesiellaceae (p = 0.041), while negative correlations were found with Prevotella (p = 0.003) and Catenibacterium (p = 0.019) ^[3][85]. Another research work reported that Fecalibacterium, reduced in the duodenal and rectal mucosal microbiota of 33 IBS-D patients compared to 15 healthy controls (4.1% vs. 1.8%, respectively), was negatively associated with abdominal bloating and stool consistency, while Hyphomicrobium, increased in the intestinal microbiota of these patients, was positively associated with abdominal pain and stool frequency ^[4][80].

Lately, attention has been given to the microbial-associated metabolites, which could represent possible biomarkers or targets for treatment, prospecting the opportunity to recognize in blood or stool samples specific disease patterns and therapeutic approaches, independently from the classifications already available, and then moving toward a personalized treatment ^[4][80].

In a study comparing IBS-D, IBS-C, and healthy controls (77 partecipants), fecal primary bile acids were significantly more represented in IBS patients, with an increase of unconjugated primary bile acids in IBS-D patients and lower SCFAs in the IBS-C group ^[26][74]. On the other hand, results from an additional study on 16 IBS-D, 15 IBS-C, and 15 healthy subjects showed an increase in primary bile acids and a decrease in secondary bile acids in IBS-D patients’ stool samples ^[27][98]. Similar but not significant differences were observed in IBS-C patients, with evidence of a significantly lower deconjugation activity in both IBS-D (p = 0.0001) and IBS-C patients (p = 0.005) compared to healthy controls ^[27][98]. Another analysis conducted on 14 patients with IBS-D, besides the evidence of higher levels of primary bile acids (p = 0.02) and lower secondary bile acids (p = 0.03), reported a significant correlation with clinical symptoms such as stool frequency and consistency compared to healthy subjects ^[28][99]. Another larger study including 345 IBS-D patients evidenced that the increase of Clostridia was associated with the rise of fecal bile acids and with a decrease in serum fibroblast growth factor 19 (FGF19) concentration, a feedback regulator of intestinal transit, possibly explaining the augmented bowel frequency and stool water content in this IBS subtype. If confirmed, this could represent a possible biomarker in IBS-D ^[29][100]. Based on a systematic review and meta-analysis of 15 studies (case-control studies, RCTs, and self-controlled studies) on the variations of fecal SCFAs in 448 IBS patients, there was a significant increase of fecal propionate in patients compared to healthy controls; particularly, there were lower levels of propionate (standard mean difference, SMD = −0.91) and butyrate (SMD = −0.53) in patients with IBS-C, while the concentration of butyrate was higher in IBS-D patients (SMD = 0.34) than in healthy subjects ^[30][101].

According to emerging evidence, other gut microbiota metabolites like amino acids, neurotransmitters, and vitamins can play a role in the pathophysiology of IBS and could be targets for treatment ^[13][102].

Gut microbiota is also involved in the production of various gases through the fermentation of carbohydrates, including carbon dioxide, hydrogen, and methane, which results in the highest prevalence detected. The intestinal tract harbours not only bacteria but also archaea, such as methanogenic species. Methanobrevibacter smithii, Methanobrevibacter oralis, Methanobacterium ruminantium, and Methanosphaera stadtmania are dominant archaeal and methane-producing species in the human gut ^[31][32][103,104]. Animal studies showed methane slows transit through the small intestine, potentially contributing to constipation ^[33][34][105,106]. A prospective, double-blind study directly comparing methane positivity on lactulose breath tests (LBTs) to the Rome I IBS classification found methane positivity had a sensitivity of 92% and specificity of 81% for correctly identifying patients with IBS-C ^[35][107]. Additionally, a meta-analysis of nine other studies including 1.277 patients demonstrated a positive association between elevated methane levels and IBS-C diagnosis ^[36][108]. Concordantly, a reduction of methane-producing micro-organisms was observed in IBS-D and IBS-M ^[3][85]. Studies have also demonstrated methane quantitatively influences the degree of constipation in LBS patients. A direct proportional relationship was observed between quantitative methane levels at LBTs and the severity of patient-reported constipation symptoms, including decreased stool frequency and harder stool consistency ^[37][109]. Additionally, methanogenesis correlated positively with the severity of bloating and flatulence (p < 0.01) as higher gas volumes induce more distension ^[38][110]. As a potential therapeutic option, a lactone form of lovastatin was observed to inhibit enzymes involved in the methanogenic pathway and reduce symptoms associated with this subclass of IBS ^[39][111].

SIBO is a condition characterized by abnormally high concentrations of bacteria in the small intestine, presenting with abdominal distension, bloating, and diarrhea, as main symptoms ^[40][41][112,113]. In clinical practice, to diagnosis this condition, culture methods have been replaced by non-invasive breath tests ^[42][114].

A meta-analysis of 25 studies including 3.192 IBS subjects and 3.320 controls found the prevalence of SIBO in IBS was 31% (95% CI 29.4–32.6; OR = 3.7, 95% CI 2.3–6.0, p = 0.001) compared to controls (16%) ^[43][115]. More specifically, to associate SIBO and archaeal methanogenesis, the American College of Gastroenterology’s clinical practice guideline introduced the concept of intestinal methanogen overgrowth (IMO) ^[44][116]. A systematic review including 17 studies and 1653 patients evaluated the relationship between methane-positive SIBO and IBS. Data showed a prevalence of 25% (95% CI 18.8–32.4); methane production was not significantly increased in overall IBS patients compared to controls (OR = 1.2, 95% CI 0.8–1.7, p = 0.37), but it was significantly more prevalent in IBS-C compared to IBS-D (OR = 3.1, 95% CI 1.7–5.6, p = 0.0001) ^[45][117]. On the other hand, a meta-analysis found the association between SIBO and IBS subtypes was strongest for IBS-D, with a prevalence of 35.5% (95% CI 32.7–40.3), or 25.2% (95% CI 22.2–28.4) for IBS-M, compared to 22.5% (95% CI 18.1–26.9) for IBS-C ^[46][118]. To fully comprehend the role of SIBO and methanogenic archea, new evidence that clarifies the mechanisms that interact between the parties involved are required.

Overall, one of the limitations of many studies is the use of diagnostic criteria that do not always correspond to those of Rome IV, also for chronological reasons. Another missing point concerns the absence of a long-term follow-up in several investigations. Moreover, the lack of standardization in the type of sample to be analyzed for microbiota analysis inevitably creates notable heterogeneity.