The choice to conduct a systematic scoping review stems from the heterogeneity of methodological approaches in available studies, the complexity of the microbiome high-level athlete relationship, the need to map existing evidence, the importance of identifying significant gaps in current literature and the utility of synthesizing evidence to guide future research. The objectives of the systematic scoping review are to identify available evidence on the relationship between training and gut microbiome in high-level athletes with particular attention to performance related biomarkers. Furthermore, the review aims to identify recurring patterns in the microbiome response to various training modalities and intensities that may influence physical performance parameters, analyse direct correlations between specific microbial profiles and quantitative indicators of athletic performance such as power, endurance, recovery time, evaluate the effectiveness of nutritional interventions aimed at modulating the microbiome for the optimization of sports performance and highlight methodological and knowledge gaps that limit the understanding of the microbiome potential in current scientific literature.

The literature search was conducted on three databases PubMed, Scopus and Web of Science, from January 2015 to September 2025. The search string was developed using MeSH terms and keywords related to three main domains: intestinal microbiome, training/sports and athletic performance (“Gastrointestinal Microbiome” OR “Gut Microbiome” OR “Gut Microbiota” OR “Intestinal Microbiota”) AND (“Exercise” OR “Physical Activity” OR “Physical Activities” OR “Sports” OR “Athletic” OR “Athletics”) AND (“Training Program*” OR “Periodization” OR “Exercise Prescription” OR “Training Load”).

The inclusion criteria were structured following the Population, Concept and Context (PCC) framework, including high-level or elite athletes (P), the gut microbiome, nutrition, and probiotics, prebiotics, postbiotics and symbiotic/postbiotic supplementation (C) and athletic performance (C). Eligible studies included original research, including randomized controlled trials (RCTs) and observational studies, as well as systematic and narrative reviews published in English language (46, 47). The exclusion criteria were equally precisely defined, including studies on non-high-level or non-elite athletes, studies with incomplete data on the microbiome, and studies on animal models only.

The selection of studies followed a three-phase process, initial screening of titles and abstracts, full-text evaluation of potentially eligible articles, and exclusion of duplicate articles. J.C. and A.F. conducted the selection process, with a third reviewer A.P. consulted to resolve any discrepancies. Data were extracted using a standardized form, including study characteristics, design, population, duration, microbiome analysis methodology, training protocols, nutritional interventions, primary and secondary outcomes, main results and limitations.

The analysis followed a narrative-descriptive approach, particularly identifying recurring patterns and methodological gaps. Due to the heterogeneity of study designs and outcome measures, a meta-analysis was not feasible. Data synthesis focused on thematic analysis and identification of convergent findings across studies.

In accordance with scoping review methodology, no formal quality assessment tool was applied to individual studies, as the primary aim was to map the existing evidence and identify knowledge gaps rather than assess methodological quality (46, 47). However, study characteristics including design, population, methodology, and limitations were systematically recorded during data extraction to provide contextual information for interpretation of findings. For the 12 experimental studies included, we assessed key methodological characteristics including study design, sample size, presence of control groups, intervention duration and dropout rates (Supplementary Table S1). This evaluation revealed that 5 studies employed RCTs (48, 50, 52, 54, 55), 5 were observational/cross-sectional studies (17, 26, 36, 51, 53), 1 used a controlled intervention design (22) and 1 was a case study (49). Sample sizes ranged from 1 to 84 participants, with intervention durations varying from 7 days to 20 weeks. Key methodological challenges identified across studies included small sample sizes in most studies, short intervention durations in several studies and varied analytical approaches limiting direct comparisons. Additionally, conducting research with elite or high-level athletes may presents intrinsic logistical and ethical constraints, including limited participant availability, competition schedules and restrictions on experimental interventions during training and competition periods, which contribute to the observed methodological limitations alongside the inherent complexities of microbiome research methodologies.

Endurance sports can increase Veillonella and Prevotella (22, 26). Kuibida et al. have highlighted that marathoners possess increased levels of Veillonella that convert lactate into energy substrates, enhancing aerobic resistance but can potentially incur states of intestinal inflammation (57). Moreover, O'Donovan et al. have highlighted that athletes of specialties such as race walking and prolonged running present high levels of Lactobacillus acidophilus (17). A detailed case study by Álvarez-Herms et al. provided interesting insights into the possible dynamic changes of the microbiota in an elite trail runner during an entire competitive season (49). The athlete presented maximal oxygen uptake (VO₂max) levels of 92 ml/min/kg and showed a progressive increase in microbial diversity compared to peak performance (49). Furthermore, a possible distinction emerged between the effects of short (42 km) and long races (172 km) for the sport discipline, where 42 km races appeared to promote an increase in presumably beneficial bacteria and SCFAs producers, including Anaerotruncus, Butyricoccus, Clostridium butyricum and especially Lachnospira, Coprococcus, while 172 km races appeared to induce transient dysbiosis with increased opportunistic pathogenic bacteria including Klebsiella, Citrobacter, Fusobacterium, Salmonella enterica and Shigella and reduction of protective commensal bacteria (49). A positive association between microbial diversity and aerobic performance emerges from Przewłócka et al. data, which detected specific correlations between changes in VO₂max and variations in Bacteroides populations. At the same time, anaerobic power was correlated with changes in Fusicatenibacter (50). Furthermore, training itself favors an increase in microbial diversity. Bielik et al. observed that high-intensity training increased the Shannon index regardless of probiotic supplementation (52).

Dedicated high-intensity training (HIT) in swimmers can generate an increase in α-diversity regardless of the use of probiotics (52). Furthermore, Kuibida et al. report that in athletes who practice power sports such as sprint specialties a prevalence of Bacteroidescoccae is observed (57). Additionally, in sports with mixed patterns between static and dynamic components, such as fencing, higher levels of Anaerostipes hadrus have been seen compared to other sports (17). Other sports with high static and high dynamic components, such as rowing, have shown pathways more expressed for folate and amino acid biosynthesis (17). Other authors, analyzing 12 elite freestyle wrestlers, highlighted how the effectiveness of pre-competition weight control may be correlated with distinct microbial patterns (51). Wrestlers with effective weight control, characterized by a difference from the target weight of less than 2 kg (<2 kg), were associated with low carbohydrate and high protein diets. Furthermore, they presented greater abundance of Solobacterium, Rothia, Fusicatenibacter, Abiotrophia, Brucella and Streptococcus, a more appropriate nutritional structure and greater adaptability to training compared to the group with less effective weight control, with a difference from the target weight greater than 2 kg (>2 kg) (51). The group with less effective weight control (>2 kg) instead showed higher levels of Christensenellaceae, Oscillospiraceae, Eubacterium siraeum and Lachnospiraceae, positively correlated with relative carbohydrate intake (51). Moreover, this group presented signs of possible inadequate adaptation to intensified training load, evidenced by the presence of leukocytes, occult blood and proteins in urine, suggesting a relationship between microbiota, metabolic adaptation and performance in wrestling (51).

Team sports can generate a mixed pattern of microbiota adaptation. Interventions like Tai Chi could be a low-intensity strategy to integrate into training as a possible microbiota modulator, generating an increase in Bacteroidetes and a decrease in Proteobacteria (54). Moreover, team sports present high levels of dynamism, such as field hockey, which has been seen to have very high levels of Lactobacillus acidophilus (17).

SCFAs produced by the gut microbiome emerge as key mediators in the relationship between microbial composition and athletic performance. Fernandez-Sanjurjo et al. have identified significant correlations between Coriobacteriaceae and acetate, between Coriobacteriaceae and isovalerate and between Bifidobacteriaceae and isobutyrate, highlighting the role of these bacterial metabolites in supporting energy metabolism during physical exercise (26). The same authors indicate that the relationship between microbiota and performance depends not on a single factor but on multiple factors.

The differential metabolomic analysis in the study by Fu et al. on 12 freestyle wrestlers identified 371 different metabolites between the group with effective weight control (<2 kg) and the group with ineffective weight control (>2 kg), with particular relevance for lipid and amino acid metabolism (51). Some key metabolites negatively correlated with carbohydrate intake included cholic acid, clofibrate, angiotensinamide and prostaglandin J2 (51). Functional enrichment revealed some specific patterns, including downregulation of linoleic acid metabolism and upregulation of tryptophan metabolism in the effective weight control group, suggesting a more favourable metabolic profile for inflammation management and energy regulation during the pre-competition phase (51).

Diet is a determining factor influencing microbiota composition and consequently athletic performance. Furber et al. have demonstrated that microbiota stability is associated with better endurance performance (48). Specifically, a high-carbohydrate diet (HCD) was able to improve endurance performance, while a high-protein diet (HPD) had the opposite effect (48). Supporting this thesis, Sanjurjo et al. indicate that carbohydrates are key factors for the positive modulation of performance and the microbiota of athletes (26). Furthermore, Murtaza et al. have highlighted that a high-carbohydrate diet (HCHO) induced maintenance of microbial diversity and improvement in race performance, as did a diet with periodized carbohydrates (PCHO), concluding that a diet low in carbohydrates and high in fats (LCHF) induced negative alterations, including reduced exercise economy, slower race performance and a reduction of Faecalibacterium spp., increase Dorea spp. and Bacteroides (22). Lastly, Jäger et al. reported that diet is a key modulator of the microbiota, where nutritional variations can generate a modification in the gut microbiota (32). Whey protein intake could positively influence microbial diversity, while greater consumption of carbohydrates and fiber in athletes is associated with an increase in Prevotella (32). The effect of fats on the microbiota of athletes is still not fully explored, their quality seems to have a determining role for athletes’ health (32).

The periodization of training and nutrition can generate cyclic alterations of the microbiota during competitions and significant differences can occur between preparation and transition phases (26, 50). In swimmers, decreases in training volume combined with an increase in intensity can induce an increase in α-diversity (52). In line with these results, Akazawa et al. have highlighted that in Japanese elite athletes, during the transition phase, there was a significant decrease in the genera Bifidobacterium, Parabacteroides and Alistipes and an increase in Prevotella (36). Furthermore, they highlighted a significant reduction in the genus Bacteroides and an increase in Blautia and Bifidobacterium during the specific preparation phase, with changes correlated to modifications in VO₂max and anaerobic power (36). Other authors indicate a correlation between microbial stability and the optimization of physical performance (48). Some authors have shown that intensified training periodization can influence the gut microbiota, with variations partly associated with the effectiveness of weight control (51). The functional prediction of the study revealed that the microbiota of the group with less effective weight control was more enriched in riboflavin metabolism, as it could potentially play a compensatory mechanism to maintain GI health homeostasis under intensified stress (51). This suggests that the microbiota could serve as a possible biomarker for training load adaptability and effectiveness of weight control strategies in weight category sports (51). Charlesson et al. also introduce an often overlooked aspect in athlete microbiome research, the importance of gut transit time as a gut health marker (53). Their results show that low training load periods could involve slower transit times, evidenced by reduced evacuation frequency and a greater number of participants experiencing absence of bowel movements (53). Increased α-diversity, reduced SCFAs concentrations and reduced Bacteroides abundance were observed when transit times were slower (53).

Some authors support the efficacy of mixed probiotic supplementation with Bifidobacterium lactis, Levilactobacillus brevis, Lactobacillus acidophilus, Bifidobacterium bifidum and Lactococcus lactis, as they have been seen to be able to improve various aspects of athletic performance associated with the use of Vitamin D3. Przewłócka et al. reported that the combination of probiotics and vitamin D3 increased the time to exhaustion during physical effort. In contrast, no significant changes were observed in the group with vitamin D3 alone (50). Jäger et al. reported a significant increase in VO₂max and aerobic power with a yogurt supplementation containing Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus for a period of 30 days (32). They also reported increased endurance performance through L. Plantarum TWK10 and a better response to training load in athletes supplemented with a multispecies probiotic blend (32). Still, other studies have not found the same effects, so the authors indicate that the regulatory mechanisms of probiotics might have an indirect action, modulating other systems (32). Lastly, the authors suggest that supplementation with multi-strain probiotics is associated with improved aerobic performance, including increased VO₂max, aerobic power and time to exhaustion (32). Yang et al. have highlighted specific effects of various probiotic strains, L. plantarum TWK10 might be able to increase muscle mass more and improve forearm grip strength, L. Plantarum PS128 has been seen to be able to improve endurance performance and reduce indices of muscle damage and L. casei Shirota might be able to improve competitive anxiety and perceived stress (59).

In the field of strength and power, Toohey et al. analyzed the effects of Bacillus subtilis in female athletes, finding significant improvements in squat 1RM, deadlift 1RM, bench press 1RM and vertical jump (55). It is necessary to indicate that these improvements were similar to the placebo group, except for the significant reduction in body fat observed only in the probiotic group. This study observed no advantages of probiotic supplementation in increasing strength and power (55). The impact of probiotics on body composition parameters and athletic performance remains uncertain, requiring further studies to clarify the role of probiotics as ergogenic supplements (32).

A crucial aspect of athletic performance, especially in intense and prolonged training contexts, is the capacity for recovery and infection resistance. Heimer et al. have highlighted that probiotics have significant positive effects on the immune system in 50% of selected studies on athletes (34). Notably, endurance athletes showed the most significant reductions in pro-inflammatory factors following single probiotic intake (34). Confirming the theory, some authors report that a high training load can compromise the immune status of athletes, thus increasing the risk of upper respiratory tract infections (URTI) (32). Furthermore, the authors have highlighted that probiotic integration benefits the prevention of URTI (32). In high-level athletes, through probiotics such as L. fermentum, L. helveticus Lafti and L. gasseri, there was a decrease in symptoms of URTI and a more excellent state of health and recovery, changes not obtained through multi-strain probiotics (32). GI problems are prevalent in endurance athletes and can compromise physical performance and nutrient absorption, generating difficulties for athletes’ health. The authors have highlighted protective effects from 27 to 33% after probiotic intake on GI symptoms, but the topic is still controversial (34). Jäger et al. also indicate that studies on the effects of probiotics in athletes show contrasting results due to methodological variability (32). However, some research reports benefits through multi-strain probiotics with B. bifidum W23, B. lactis W51, E. faecium W54, L. acidophilus W22, L. brevis W63 and L. lactis W58 on reducing zonulin levels in high-level athletes, but further studies are needed regarding the reduction of GI symptoms and probiotic intake (32).

Teglas and Radak provided a detailed overview of probiotic effects in different sports, emphasizing how supplementation effectiveness depends on multiple factors, including specific strain, colony forming units (CFU), duration and frequency of intake (56). These factors can modulate the physiological impact of probiotics both independently from training protocols and in interaction with them, producing different effects. In endurance sports, probiotic supplementation appears to positively influence lipid metabolites, including SCFAs, modulate VO₂max and improve exercise duration, while in sports characterized by intermittent exercise, probiotics may reduce inflammatory process activity and improve factors related to psychological stress, such as anxiety and depression (56).

In runners, administration of Bifidobacterium animalis Lactis and Lactobacillus acidophilus (10 × 10⁹ CFU for 30 days) appeared to reduce pro-inflammatory cytokine production and maintain CD8 + cell and effector memory cell populations (56). The use of Pediococcus acidilactici and Lactobacillus plantarum (3 × 10⁹ CFU for 4 weeks) did not produce effects on GI symptoms (56). Bifidobacterium longum longum OLP-01 (1.5 × 10¹⁰ CFU for 5 weeks) showed increased distance covered and greater intestinal microbiota abundance (56). Bifidobacterium lactis, Lactobacillus brevis, Lactobacillus casei, Lactococcus lactis, Lactobacillus acidophilus, Bifidobacterium bifidum and Ligilactobacillus salivarius (2.5 × 10⁹ CFU for 12 weeks) improved VO₂max, 60 s ventilation, functional capacity, respiratory reserve and exercise capacity, also reducing GI symptoms (56). Finally, Lactobacillus helveticus Lafti L10 (5 × 10⁹ CFU for 6 weeks) appeared to reduce time to exhaustion in runners (56).

In road cyclists, supplementation with a multi-strain combination of Lactobacillus and Bifidobacterium (1 × 10¹¹ CFU for periods of 4, 12 and 16 weeks) showed increased aerobic capacity, VO₂max, exercise duration to exhaustion and reduced heart rate (56). Similarly to cyclists, a combination of Lactobacillus, Bifidobacterium, Enterococcus and Bacillus strains for 90 days reduced GI symptoms without generating effects on VO₂max and time to exhaustion (56). In skiing, Bifidobacterium lactis BL-99 (1 × 10⁹ CFU for 8 weeks) appeared to increase levels of SCFAs and polyunsaturated fatty acids, bile salts, knee extensor strength and VO₂max (56).

In team sports, probiotic effects are equally varied. Lactobacillus casei Shirota (3 × 10¹⁰ CFU for 8 weeks) appeared to improve reaction time in cognitive tests of soccer players, while a blend of Lactobacilli, Bifidobacteria and Streptococcus strains (4.5 × 10¹¹ CFU for 4 weeks) showed no effects on VO₂max (56). Moreover, Lactobacillus casei, Bifidobacterium lactis V9 and Lactobacillus plantarum P-8 (≥6–8 × 10⁹ CFU for 6 weeks) showed URTI reduction, increased Secretory Immunoglobulin A levels, decreased inflammatory factors, reduced maximum heart rate and lactate elimination rate (56). Administration of Bifidobacterium lactis CBP-001010, Lactobacillus rhamnosus CNCM I-4036, and Bifidobacterium longum ES1 (≥1 × 10⁹ CFU for 1 month) reported reduced stress, anxiety, and depression and increased post-exercise dopamine (56). Administration of Lactobacillus casei Shirota (3 × 10¹⁰ CFU for 6 weeks) in badminton players showed stress reduction and increased aerobic capacity, while Bifidobacteria and Lactobacilli strains (≥1.25 × 10¹⁰ CFU for 23 days) in basketball players appeared to show reduction of inflammatory processes and apoptosis of peripheral lymphocytes (56).

Pierudzka et al. highlighted an important gap in the scientific literature regarding the interaction between hormonal contraceptives, gut microbiota and exercise adaptations in female athletes (58). Only one pilot study appears to have directly investigated this relationship, showing that hormonal contraceptive use in physically active women appears to be associated with a reduction in SCFAs producing taxa, suggesting potential implications for energy metabolism and exercise adaptations (58). This complex interaction requires further research to develop personalized nutritional and training strategies for high-level female athletes using hormonal contraceptives.

Training periodization emerges as a key factor in microbiome modulation. Longitudinal studies have highlighted cyclic alterations in microbial composition during different preparation phases, with significant variations between load and recovery periods (26, 50, 52). Charlesson et al. provided new considerations on the impact of training load on intestinal health markers in elite athletes (53). Their longitudinal study on elite rowers showed that high training load periods, compared to low training load periods, can modify SCFAs concentrations, reduce evacuation frequency, increase Bacteroides abundance and decrease α-diversity, with a lower Firmicutes/Bacteroides ratio (53). The authors also identified that training load independently influences microbiome composition even when controlling for changes in diet quality, evacuation frequency and sex (53). In particular, diet quality explains part of the microbial variation during high training load periods, while during low training load periods the training stress score appears to contribute more to the observable microbial variation (53) (Figure 2).

Nutritional interventions show a significant impact on microbiome composition and consequently on performance. Probiotic supplementation with multi-strains or single strains has been shown to generate benefits on aerobic physical performance and intestinal health, while on strength performance, at the moment, no significant benefits have been measured (32, 50, 55). High carbohydrate diet strategies favour microbial profiles associated with better endurance performance, while low carbohydrate diets can negatively alter the microbial ecosystem, as can high protein diets change the diversity of the microbiota in athletes (18, 22) (Figure 2).

This systematic scoping review was conducted following the rigorous PRISMA-ScR methodological framework and was registered a priori with the Open Science Framework. The literature search was comprehensive, utilizing three major databases: PubMed, Scopus and Web of Science with a well defined search strategy. The inclusion of both experimental studies and systematic/narrative reviews provided a comprehensive overview of the current evidence. The narrative/descriptive analysis approach was appropriate for identifying patterns and knowledge gaps in this emerging field. Despite the robust methodology employed in this systematic scoping review, it is essential to highlight some significant methodological limitations that characterize this field of research. The variability in microbiome analysis methodologies across different studies limits the direct comparability of results, and the often reduced sample sizes of studies decrease the statistical power and generalizability of results (60). Furthermore, most studies have included male athletes. The interactions between diet, training, microbiome and performance make it difficult to isolate the specific contribution of single factors. The high variability in microbiome composition between individuals, moreover, represents a confounding factor that complicates the interpretation of results. The variability in intervention durations and training protocols is another limitation. Finally, there is an evident lack of standardization in outcome measures across studies.

Despite significant advances in gut microbiome research, science still faces the challenge of fully understanding the nuances of individualized responses to interventions, species redundancy, the importance of strain-specific variations, uncharacterizable microorganisms, host microbiome interactions and the long-term effects of microbial manipulations. Although these knowledge gaps, the collected evidence suggests several practical applications and directions for future research, but the fundamental pillar for microbiome modulation remains the nutritional approach (61). Yang et al. emphasize the need to consider individual differences in athletes and precision nutrition as a crucial element to optimize the effects of the microbiome (59). Probiotic supplementation is highly variable in effects but also essential and reasoning with personalized approaches with particular attention to strain specificity based on the athlete's individual profile, considering the sports discipline practiced, baseline microbiome characteristics, specific physical performance objectives and the athlete's health status, could in the future play a significant role in athletic performance. Some authors provide evidence supporting the need to synchronize dietary strategies with training and competition phases to maintain an optimal microbiota state (26, 48). Supporting these considerations, Jarret et al. report that current evidence indicates gut microbiome modulation by means of probiotic supplementation can have ergogenic benefits on endurance performance, while preliminary findings also suggest potential benefits for strength and power performance, although further research is required (62). The mechanisms are not fully understood, but may involve improved exercise recovery, immune function, nutrient absorption and alleviation of GI symptoms (62). These findings reinforce the importance of the previously mentioned synchronization approach, where it would be essential to consider diets according to the training period, adapting the nutritional approach to optimize microbial composition based on the specific needs of different phases of training periodization. The results support the utility of microbiome monitoring as a potential biomarker of training, recovery and health status in high-level athletes, integrating microbiome analysis in training periodization. Moreover, standardization in processual methodologies is necessary since, today it remains a very complex topic (63).