Exploring the Dynamic Relationship between the Gut Microbiome and Body Composition across the Human Lifespan: A Systematic Review

This systematic review aimed to identify different gut microbiome profiles across the human lifespan and to correlate such profiles with the body composition. PubMed, Scopus, and Cochrane were searched from inception to March 2022. Sixty studies were included in this systematic review. Overall, the gut microbiome composition in overweight participants exhibited decreased α-diversity, decreased levels of the phylum Bacteroidetes and its taxa, and increased levels of the phylum Firmicutes, its taxa, and the Firmicutes/Bacteroidetes ratio, in comparison to normal-weight participants. Other body composition parameters showed similar correlations. Fat mass and waist circumference were found to correlate positively with the Firmicutes taxa and negatively with the Bacteroidetes taxa. In contrast, lean body mass and muscle mass demonstrated a positive correlation with the Bacteroidetes taxa. Notably, these correlations were more pronounced in athletes than in obese and normal-weight individuals. The composition of the gut microbiome is evidently different in overweight individuals or athletes of all age groups, with the former tending towards decreased Bacteroidetes taxa and increased Firmicutes taxa, while a reversed relationship is observed concerning athletes. Further studies are needed to explore the dynamic relationship between energy intake, body composition, and the gut microbiome across the human lifespan.


Introduction
The gut microbiome is involved in multiple essential functions responsible for the normal functioning of the intestine and the host [1,2], but its composition is unique to each person.In fact, there is less than 10% similarity between any two individuals [3].Its formation is determined early from birth through adulthood and modified by genetic and environmental factors, such as diet, physical activity, age, gender, sleep, smoking, and antibiotics [1,4].
The brain participates dynamically in energy balance regulation via its ability to communicate with the peripheral organs through various nerve and chemical signals, most of them coming from the gastrointestinal tract, a relationship called the gut-brain axis [5].The activation of neuropeptide Y/agouti-related peptide (NPY/AGRP) neurons in the hypothalamus of the brain by the hormone ghrelin has an orexigenic effect by stimulating an increase in appetite and a decrease in energy expenditure [6,7].Ghrelin is negatively correlated with the genera Bifidobacterium, Lactobacillus, Blautia coccoides, and Eubacterium rectale and positively correlated with the genera Bacteroides and Prevotella [8,9].The activation of pro-opiomelanocortin/cocaine-amphetamine-related transcript (POMC/CART) and the suppression of NPY/AGRP neurons by the hormones insulin, leptin, cholecystokinin 2 of 36 (CCK), peptide YY (PYY), glucagon-like-peptide 1 (GLP-1), and oxyntomodulin (OXM) leads to the opposite, anorexigenic effects [6,7].
Moreover, the specific pathways through which the gut microbiome communicates with the brain and interacts with energy expenditure, body weight, and body composition are well known.The first mechanism involves lipopolysaccharides (LPS), found in the cell walls of Gram-negative bacteria on macrophages and adipose tissue.LPS activate a cascade of pro-inflammatory responses that are accountable for a chronic state of underlying inflammation [8][9][10][11][12][13].The second mechanism involves short-chain fatty acids (SCFAs), which are produced by the fermentation of fiber.In addition to contributing to approximately 10% of energy intake, they participate in other metabolic pathways, such as promoting hepatic lipogenesis and gluconeogenesis, inflammatory reduction, and an increase in GLP-1 and PYY production [8,[10][11][12].The latter mechanism involves bile acids, which are fermented by the colon microbiome to produce secondary bile acids.Secondary bile acids later promote increased energy expenditure and the production of GLP-1 [8].
As a result, the gut microbiome profile appears to be different in overweight and obese compared to normal-weight individuals, demonstrated through various studies [14][15][16].Dysbiosis, or the imbalance of the gut microbiota, has been associated with inflammatory responses observed in clinical conditions, underlying the microbiota's influence on gastrointestinal health and disease mechanisms [17].Recent findings underscore the potential of monitoring the gut microbiome for diagnostic and therapeutic strategies for inflammatory bowel diseases (IBD), irritable bowel syndrome (IBS), and colorectal cancer, highlighting the microbiota's integral role in gastrointestinal health [18].Notably, it is suggested that gut microbiome modification could be a potential strategy for the early prevention and treatment of relevant conditions and obesity, through improving dietary habits; taking probiotics, prebiotics, and synbiotics; and fecal microbiota transplantation from healthy individuals [10,19,20].
However, based on the literature review, the formation of the gut microbiome in relationship to the body composition is poorly systematized.Specifically, there are no summarized gut microbiome profiles across the human lifespan according to age groups in healthy individuals.Thus, the current systematic review focused on identifying different gut microbiome profiles in healthy individuals, from children to older adults, and to correlate such profiles with the body composition.

Information Sources and Search Strategy
This systematic review was based on the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [21].The literature search of the systematic review was carried out on 18 March 2022 in the PubMed, Scopus, and Cochrane databases using the following keywords: (((healthy individual* OR human* OR obes*) NOT (disease* OR disorder* OR syndrome OR diabetes OR cancer)) AND ("gut microbio*" OR "intestinal microbio*" OR "fecal microbio*" OR "cecal microbio*" OR microflora OR "gut bacteria" OR "intestinal bacteria")) AND ("body composition" OR "fat-free mass" OR "fat mass" OR "body fat" OR "body mass" OR "body mass index" OR BMI OR "energy expenditure" OR "basal metabolic rate" OR BMR OR "resting metabolic rate" OR RMR).A supplementary search for relevant studies was conducted from the reference lists of the screened manuscripts.

Eligibility Criteria
The research question and inclusion and exclusion criteria were determined using the PICO strategy (Patient, Intervention, Controls, Outcome).The inclusion criteria were (1) primary research; (2) studies that presented the gut microbiome in the large intestine; (3) studies written in the English language; (4) studies that had as a population healthy children, adults, older adults, and postmenopausal women; (5) studies that intervened by providing probiotic, prebiotic, and symbiotic supplements; (6) studies that performed an intervention by modifying the diet or physical activity or both.The exclusion criteria were (1) non-primary research (i.e., reviews and case studies); (2) studies not written in English; (3) studies with a non-healthy population (except obese); (4) studies that presented the gut microbiome in other areas, such as the mouth; (5) studies that involved twins, infants, pregnancy, or breastfeeding; (6) studies that performed an intervention by providing medication.The PICO criteria for the inclusion and exclusion of studies are presented in Table 1.

Data Collection Process
Primary screening was conducted by two independent researchers (I.K., A.V.) using Microsoft Excel, according to the established eligibility criteria.Full-text secondary screening for the selection of the final articles was also conducted by these two independent researchers, while, where there were conflicts, a third independent researcher (CDG) resolved them.

Data Extraction and Quality Assessment
Data extraction from the final articles was conducted by one researcher (I.K.), who presented the results in four tables based on age groups (children, adults, older adults, and whole age range).Extracted data included the name of the first author and publication date, sample size, gender, age, BMI category, body composition, and results.
The quality assessment was conducted by one researcher using the Newcastle-Ottawa Scale (NOS) tool, adapted according to the study design [22].The NOS tool consists of three sections regarding sample selection, a search for confounding factors, and study outcomes.As confounding factors, in the present systematic review, a check for the exclusion of antibiotic and/or probiotic intake was determined.The tool involves eight or nine questions and ten is the maximum score achieved.Due to the final score, studies were classified as "low quality" if the score was <5, as "moderate quality" if the score was 5-7, and as "high quality" if the score was >7.
Extracted data were categorized into four tables based on age groups: (i) children, <18 years, (ii) adults 18-65 years, (iii) older adults >65 years, and (iv) whole age range, children to older adults.The extracted data in each table were further categorized according to (1) sex-males and females; (2) BMI category-children were classified according to growth charts and adults according to BMI into (i) normo-weight, (ii) overweight, and (iii) obese, also considering the ethnicity-specific BMI cutoffs, as provided by the original articles; (3) body composition-some studies included information regarding body composition besides BMI, such as body fat percentage and muscle mass; (4) athletes-athletes were included in some studies as part of the sample to observe differences between them and non-athlete individuals.Athletes' competition levels were determined based on the description provided by each study.

Study Selection
During the search process using the keywords in the PubMed, Scopus, and Cochrane databases, 995 potentially relevant studies were found, with 614 studies remaining after duplicates were removed and the filters "human" and "humans" were applied.After the primary screening, which included the title and abstract reading, 188 studies were selected for full-text screening.The final studies that met the eligibility criteria and were included in this systematic review totaled 60. Figure 1 shows the study selection process in detail, according to the PRISMA 2020 guidelines.
to growth charts and adults according to BMI into (i) normo-weight, (ii) overweight, and (iii) obese, also considering the ethnicity-specific BMI cutoffs, as provided by the original articles; (3) body composition-some studies included information regarding body composition besides BMI, such as body fat percentage and muscle mass; (4) athletes-athletes were included in some studies as part of the sample to observe differences between them and non-athlete individuals.Athletes' competition levels were determined based on the description provided by each study.

Study Selection
During the search process using the keywords in the PubMed, Scopus, and Cochrane databases, 995 potentially relevant studies were found, with 614 studies remaining after duplicates were removed and the filters "human" and "humans" were applied.After the primary screening, which included the title and abstract reading, 188 studies were selected for full-text screening.The final studies that met the eligibility criteria and were included in this systematic review totaled 60. Figure 1 shows the study selection process in detail, according to the PRISMA 2020 guidelines.
The 18 studies with children as the target group were published between 2010 and 2022, with sample sizes ranging from 21 to 502 participants, while countries worldwide were included.None of the 18 studies included athletes.Fourteen studies were cross-sectional, one was a case-control study, and three were clinical trials.The 32 studies with adults as the target group were published between 2011 and 2021, with sample sizes ranging from 20 to 263 participants, while countries from all over the world were included.Three studies included athletes in their samples [34,39,64].Nineteen studies were cross-sectional, one was longitudinal, 11 were clinical trials, and one was a comparative study.The four studies with older adults as the target group were published between 2017 and 2021, with sample sizes ranging from 22 to 201, with participants from Japan, Slovakia, and Italy.One study of this category included athletes in its sample [46].Three studies were cross-sectional and one was a clinical trial.The six studies with participants from 14 to 88 years old were published between 2010 and 2021, with sample sizes ranging from 26 to 528, with participants from European and Asian countries.One study included athletes in its sample [49] and all six were cross-sectional.
The gut microbiome analysis was presented in each age group in terms of αdiversity, β-diversity, and bacterial taxonomy, including dominant phyla and genera and correlations of the gut microbiome composition with the BMI and other body composition measurements.

Children
Most of the results from the studies in this category were presented comparing normalweight with underweight or overweight children (Table 2).Regarding α-diversity, seven studies took it into consideration [23,24,[26][27][28][29]57]; however, only two studies reported statistically significant differences (p < 0.05) in relation to body composition [23,28].The first study identified three groups according to muscle mass [23], and the second one showed less α-diversity in obese children compared with normal-weight children [28].Only one study included β-diversity, with statistically significant differences between obese and normal-weight children (p < 0.05) [28].
Other body composition parameters, besides BMI were correlated with the gut microbiome composition in five studies [23,31,55,58,69].More specifically, a positive correlation was observed between the phylum Firmicutes and the circumferences of the waist and head [58], while correlations were also observed between Bacteroidetes (p = 0.031; p = 0.012; p = 0.003), the F/B ratio (p = 0.075; p = 0.032; p = 0.002), Actinobacteria (p = 0.039; p = 0.053; p = 0.078), and visceral, subcutaneous, and hepatic fat [69].A positive correlation was observed between the family Lactobacillaceae and visceral fat [55], while a correlation was also observed between the family Ruminococcaceae and the fat-free mass index (FFMI) Z-score in boys (p = 0.027) [31].At the genus level, Faecalibacterium and Lachnospira were positively correlated with at least one of the following: ratio of total body lean soft tissue mass (TSM) to weight (TSMR), appendicular skeletal mass (ASM), ratio of appendicular skeletal mass to height (ASMI), and ASMI z-score.They were negatively correlated with at least one of the following: ratio of TSMR, total body lean soft tissue mass/total body fat (TSM/TBF), appendicular skeletal mass to weight (ASMR), appendicular skeletal mass/appendicular fat mass (ASM/AFM), and ASMR z-score [23].The genera Actinomyces, Bifidobacterium, Streptococcus, and Blautia were positively correlated with body fat storage, while, in contrast, the genera Odoribacter, Oscillospira, Bacteroides, and Faecalibacterium were negatively correlated with fat [69].NW-Obese: no significant differences from phyla to genus, F/B ratio, richness, alpha diversity.

Adults
The results from the studies in the adult category were presented by comparing normal-weight adults with either athletes or overweight/obese adults (Table 3).Five studies showed statistically significant differences in α-diversity (p < 0.05) [34,40,71,73,80], while two studies showed no differences (p > 0.05) [33,79].More specifically, α-diversity was significantly lower in overweight/obese individuals, compared with the normal-weight control group (p < 0.05) [40,71,73], although Kasai and his colleagues reported the opposite result [80].Clarke et al. [34] compared elite athletes with two groups of non-athletes, including both low and high BMI levels.Elite athletes showed statistically significantly higher levels of α-diversity compared with both groups, while no differences between the control groups were observed.Statistically significant differences between normal-weight and overweight individuals for β-diversity (p < 0.05) were observed in two studies [33,40].
Males compared with Females → ↑ Bacteroidetes (p < 0.001), ↓ Firmicutes (p < 0.001), Actinobacteria (p = 0.04).Other body composition parameters, apart from BMI, were correlated with the gut microbiome composition in seven studies [34,40,41,45,65,66,70].Obesity parameters, examples of which include the body fat percentage, visceral fat, waist circumference, and waist/hip ratio, were positively correlated with Firmicutes, the Firmicutes taxa, and the F/B ratio and negatively correlated with Bacteroidetes and the Bacteroidetes taxa, both in males and females (p < 0.05) [40,70].A positive correlation was also observed in women between the body fat percentage and Eubacterium rectale and Clostridium coccoides [45].On the other hand, another study showed a negative correlation between the waist circumference and Clostridium leptum (p < 0.05), as well as between body fat and Bifidobacterium, Clostridium leptum, and Lactobacillus plantarum in women (p < 0.05) [66].A study conducted with a male sample showed a negative correlation between body fat and Faecalibacterium (p < 0.05) [41].Both lean body mass and fat mass were negatively correlated with the Firmicutes taxa in males and females and positively correlated with Faecalibacerium in males (p < 0.05) [40,41].Clarke et al. [34] compared elite male athletes with a control group of men with high BMI levels, who were not statistically significantly different from athletes.The two groups differed in the body fat percentage, lean body mass, and waist/hip ratio.Statistically significantly higher levels of the Akkermansiaceae family (p = 0.049) and the Akkermansia genus (p = 0.035) and lower levels of the Bacteroidetes phylum (p = 0.022) were observed in athletes [34].Finally, one study did not show any statistically significant correlation between the gut microbiome composition and lean body mass or fat mass (p > 0.05) [65].

Older Adults
The results from the studies in the older adult category were presented by comparing older adults with different BMIs and long-term athletes versus sedentary control groups (Table 4).Two studies investigated α-diversity [46,48]; one did not show any statistically significant difference between the athlete group and the sedentary control group (p > 0.05) [46], while the other study categorized its sample into three groups according to different gut microbiome compositions and did show significant differences between groups (p < 0.05) [48].None of the studies investigated β-diversity.

Whole Age Range
The results from the studies in the whole age range category were presented by comparing people with different body composition measurements, regardless of age (Table 5).Three studies showed statistically significant differences for α-diversity (p < 0.05) [49,51,52].The α-diversity was higher in athletes compared with non-athletes (p < 0.05) [49], older adults compared with adults (p < 0.05) [51], and normo-weight compared to obese individuals (p < 0.05) [52].None of the studies investigated β-diversity.
Correlations of the gut microbiome with other body composition parameters, besides BMI, were found only by Kulecka et al. [49].The sample was categorized into three groups, marathon runners, skier athletes, and a sedentary control group.The body composition parameters, like body fat, lean body mass, and muscle mass, differed between the two athlete groups and the control group (p < 0.05).The results showed reduced levels of Bacteroides and increased levels of Prevotella in both athlete groups compared to the control group (p < 0.05).Increased levels of the F/B ratio were also observed in skiers compared with the control group (p = 0.043), while no statistically significant difference was observed in marathon runners (p > 0.05).
The main differences in the gut microbiome composition in all BMI categories in all age groups are presented in Table 6.Figures 2 and 3 show a comparative representation of the gut microbiome's formation across the human lifespan.Children, adults, and older adults are categorized according to BMI into (i) normo-weight, (ii) overweight, (iii) obese, and (iv) athletes and are compared in terms of the gut microbiome composition regarding α-diversity and the most commonly found phyla, genera, and species.
Correlations of the gut microbiome with other body composition parameters, besides BMI, were found only by Kulecka et al. [49].The sample was categorized into three groups, marathon runners, skier athletes, and a sedentary control group.The body composition parameters, like body fat, lean body mass, and muscle mass, differed between the two athlete groups and the control group (p < 0.05).The results showed reduced levels of Bacteroides and increased levels of Prevotella in both athlete groups compared to the control group (p < 0.05).Increased levels of the F/B ratio were also observed in skiers compared with the control group (p = 0.043), while no statistically significant difference was observed in marathon runners (p > 0.05).
The main differences in the gut microbiome composition in all BMI categories in all age groups are presented in Table 6.Figures 2 and 3 show a comparative representation of the gut microbiome's formation across the human lifespan.Children, adults, and older adults are categorized according to BMI into (i) normo-weight, (ii) overweight, (iii) obese, and (iv) athletes and are compared in terms of the gut microbiome composition regarding α-diversity and the most commonly found phyla, genera, and species.

Discussion
The present systematic review aimed to identify different gut microbiome profiles in healthy individuals, from children to older adults, and to correlate them with body composition formation.It was found that there are significant differences in the gut microbiome composition in individuals with excess weight or athletes across different age groups.
It was observed that the gut microbiome composition of overweight and obese participants was characterized by decreased α-diversity, mostly in adults compared to children, where only two [23,28] out of the seven studies [23,24,[26][27][28][29]57] showed statistically significant differences.In addition, decreased levels of the Bacteroidetes phylum and its taxa and increased levels of the Firmicutes phylum, its taxa, and the F/B ratio were observed in comparison to normal-weight participants.Other body composition parameters, apart from the BMI, followed similar correlations.More specifically, a positive correlation between the Firmicutes phylum, its taxa, and obesity parameters, examples of which include the body fat mass and waist circumference, was observed, while a negative correlation was observed between the Bacteroidetes phylum, its taxa, and obesity parameters.On the other hand, the Bacteroidetes phylum and its taxa were also positively correlated with the lean body mass and muscle mass.These outcomes appeared to be more significant in athletes, even compared to normal-weight individuals.
The relationship between the gut microbiome composition and body weight has recently been discovered and continues to be studied widely, especially during the last decade [10,82].Studies conducted in mice observed an alteration in body weight after a fecal transplant intervention from obese mice to mice without any microbiome; such an observation is responsible for the expanding studies conducted in humans [83].The three main mechanisms through which the gut microbiome contributes to body weight are well known and have already been described in the Introduction of the current systematic review.Briefly, the first mechanism involves LPS promoting underlying inflammation, a common sign of obesity.The second mechanism involves the SCFAs that metabolize undigested food components like fiber, resulting in 10% more energy intake, while, in contrast, they contribute to other metabolic pathways, activating the secretion of anorexic hormones.The last mechanism involves bile acids, through which energy expenditure and the secretion of anorexigenic GLP-1 are promoted [8].Despite the fact that the above mechanisms are well studied, the responsible bacteria are not yet fully identified [84].
According to the existing literature, the results for α-diversity between individuals with normal and excess weight are controversial.A meta-analysis conducted by Walters et al. [85] in 2014 did not show any statistically significant difference between normo-weight and overweight adults.In contrast, two more recent meta-analyses confirmed the reduced α-diversity in obesity observed in the current systematic review, although only two of the ten studies in Sze and Schloss's meta-analysis showed statistically significant differences (p < 0.05) [86,87].It is noteworthy that α-diversity is related to the better functionality of the gut microbiome; thus, a reduced α-diversity can lead to the disruption of the gut microbiome's functioning and, ultimately, host dysbiosis [88].Two recent systematic reviews examined the impact of exercise on α-diversity, confirming a positive association between α-diversity and individuals with high levels of fitness or cardiorespiratory fitness, as well as individuals with lower fitness levels after the impact of exercise [89,90].
At the phylum level, the F/B ratio, in the majority of the studies, was observed to be higher in obese compared with normo-weight individuals, in all age groups.However, two meta-analyses were in disagreement with our results, showing that the F/B ratio did not display statistically significant differences (p > 0.05) [15,91].Thus, this measure cannot be considered a strong indicator for the separation of individuals based on BMI [87,92].The phyla Firmicutes and Bacteroidetes are well known as the dominant phyla of the gut microbiome, making up over 90% of its composition [93].The increased levels of the Firmicutes and decreased levels of Bacteroidetes observed in obese participants in the present systematic review are in agreement with a number of studies confirming the respective relationship [14,15,92].More specifically, the phylum Firmicutes is positively correlated with parameters related to obesity, such as the body fat percentage, and negatively correlated with the lean body mass.In contrast, the phylum Bacteroidetes is negatively correlated with obesity parameters, a result that is also consistent with the present findings [19,85,94].The observed relationship between Firmicutes and obesity parameters seems to be explained by the fact that many enzymes involved in carbohydrate metabolism belong to this phylum.The exact mechanism that promotes obesity is probably the one involving the production of SCFAs, as a positive correlation has been observed between the phylum Firmicutes and SCFAs in feces.This indicates that obese individuals prevail in the fermentation of undigested nutrients in the large intestine and, by extension, in the 10% excess energy production and in body weight gain [92,[95][96][97][98].Moreover, a second mechanism concerning SCFAs can explain the observed positive correlation between the phylum Firmicutes and body fat.The fermentation of fiber by SCFAs can also lead to the promotion of hepatic lipogenesis, increasing the storage and accumulation of fatty acids and triglycerides in the adipose tissue.Acetic acid is considered to be the main culprit responsible for this process and is mainly produced by bacteria belonging to Firmicutes [19,93].
Recent meta-analyses that investigated the gut microbiome's composition in normoweight and obese individuals confirm the results of the current systematic review at the genus level.Some commonly detected genera in obese individuals are increased levels of Dorea, Eubacterium, Megasphaera, Dialister, Lactobacillus, and Streptococcus (phylum Firmicutes) and decreased levels of Bacteroides, Alistipes (phylum Bacteroidetes), Bifidobacterium (phylum Actinobacteria), Faecalibacterium, and Oscilibacter (phylum Firmicutes).However, it is obvious that the relationship between the phylum level and obesity does not necessarily expand at the genus level.For instance, the genera Faecalibacterium and Oscilibacter are reduced in obese people, while the expected observation would be increased levels due to belonging to the phylum Firmicutes [15,85,99,100].The exact mechanism through which some bacteria affect body weight has already been discovered.The bacteria Lactobacillus plantarum, Faecalibacterium prausnitzii, and Akkermansia muciniphila appear to be reduced in obese compared to normo-weight people, a correlation that was also found in the present study.The genus Lactobacillus, as a member of the phylum Firmicutes, has been associated with obesity and is found to be increased in those with excess weight.Some specific species, like Lactobacillus plantarum, have been shown to prevent dysbiosis through the production of bacteriocins that prevent the growth of pathogenic microorganisms [101].Faecalibacterium prausnitzii causes the production of butyric acid from the fermentation of undigested nutrients and is also characterized by its anti-inflammatory role, explaining its protective role against obesity [102].Akkermansia muciniphila participates in mucus metabolism and the maintenance of intestinal barrier integrity in the host, while it prevents the colonization of pathogenic microorganisms and dysbiosis [103].
As in every research study, there are some issues that need to be considered when interpreting the data of this review.Firstly, the majority of the studies included were cross-sectional; hence, their results do not reflect a cause-effect relationship.It is important to note that the prospective studies and clinical trials reported in this review included baseline data, before any intervention took place.Moreover, the heterogeneity between studies should also be considered, not only regarding the definition of obesity, which differs by country and age, but also regarding the level of bacterial taxonomy investigated by each study, making the comparison of the results difficult.

Conclusions
To conclude, the composition of the gut microbiome is evidently different in overweight individuals or athletes in all age groups.The composition of the gut microbiome in obese people comprises decreased α-diversity, decreased levels of the phylum Bacteroidetes and its taxa, and increased levels of the phylum Firmicutes, its taxa, and the F/B ratio.Besides the BMI, obesity parameters, like body fat mass, are positively correlated with the Firmicutes taxa and negatively correlated with the Bacteroidetes taxa, and lean fat mass and

Figure 1 .
Figure 1.The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for study selection.

Figure 1 .
Figure 1.The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for study selection.
AFM = Appendicular Fat Mass; ASM = Appendicular Skeletal Mass; ASMI = Ratio of ASM to Height; ASMR = Ratio of ASM to Weight; BF = Body Fat; BMI = Body Mass Index; F = Female; HMM = High Muscle Mass; LMM = Low Muscle Mass; M = Male; MMM = Medium Muscle Mass; NC = Neck Circumference; NR = Not Reported; NW = Normal-Weight; TBF = Total Body Fat; TFM = Total Fat Mass; TSM = Total Body Lean Soft Tissue Mass; TSMI = Ratio of TSM to Height; TSMR = Ratio of TSM to Weight; WC = Waist Circumference; WHO = World Health Organization; WtHR = Waist-to-Height Ratio.

Figure 2 .
Figure 2. Predominant bacterial phyla and genera across distinct age groups.Figure 2. Predominant bacterial phyla and genera across distinct age groups.

Figure 2 .
Figure 2. Predominant bacterial phyla and genera across distinct age groups.Figure 2. Predominant bacterial phyla and genera across distinct age groups.

Figure 3 .
Figure 3. Bacterial phyla and general dynamics: contrasts between obesity and normal weight and across age groups including children, adults, and athletes.Figure 3. Bacterial phyla and general dynamics: contrasts between obesity and normal weight and across age groups including children, adults, and athletes.

Figure 3 .
Figure 3. Bacterial phyla and general dynamics: contrasts between obesity and normal weight and across age groups including children, adults, and athletes.Figure 3. Bacterial phyla and general dynamics: contrasts between obesity and normal weight and across age groups including children, adults, and athletes.

Table 1 .
The PICO criteria for inclusion and exclusion of studies.

Table 2 .
Characteristics of studies investigating the gut microbiome composition in children.

Table 3 .
Characteristics of studies investigating the gut microbiome composition in adults.

Table 4 .
Characteristics of studies investigating the gut microbiome composition in older adults.

Table 6 .
Main differences in gut microbiome composition in all BMI categories and age groups.