Targeted and Untargeted Urinary Metabolomics in Exercise Science: A Systematic Review

Article information

Exerc Sci. 2025;34(2):95-107

Publication date (electronic) : 2025 May 30

doi : https://doi.org/10.15857/ksep.2025.00269

Zi-Xiao Yu, MS ¹

, Ji-Suk Chang, PhD ²

, Ah-Ram Kim, PhD ³

, Ho-Seong Lee, PhD ¹^,³^,⁴^,⁵^,

¹Department of Exercise and Medical Science, Graduate School, Dankook University, Cheonan, Korea

²Department of Sport Management, College of Sports Science, Dankook University, Cheonan, Korea

³Department of Sports Medicine, Graduate School of Sport Science, Dankook University, Cheonan, Korea

⁴Department of International Sports Studies, College of Sports Science, Dankook University, Cheonan, Korea

⁵Institute of MEDI-Sports, Dankook University, Cheonan, Korea

Corresponding author: Ho-Seong Lee Tel +82-41-550-3838 Fax +82-41-550-3838 E-mail hoseh28@dankook.ac.kr

Received 2025 April 29; Revised 2025 May 2; Accepted 2025 May 7.

Abstract

PURPOSE

Physical activity induces metabolic changes in the body. However, the biological mechanisms underlying these metabolic changes remain unclear. Metabolomics has advanced the field of exercise science, although most studies use invasive procedures for sample collection and have low rates of participant compliance.

METHODS

A systematic literature search was conducted between March 30 and April 5, 2024, to identify articles published between January 1, 2004 and January 1, 2024, across four academic databases: Web of Science, Google Scholar, PubMed, and Scopus.

RESULTS

This review analyzed 34 original research articles on urinary metabolomics in exercise science. NMR and mass spectrometry platforms were used in similar proportions of the studies, accounting for 41.2% (14/34) and 35.3% (12/34) of the studies, respectively. Untargeted approaches dominated research, accounting for 85.3% of studies (n=29), whereas targeted methodologies were exclusively used in only 14.7% (n=5).

CONCLUSIONS

Urinary metabolomics is rapidly emerging as a valuable tool in the field of exercise science. Untargeted approaches have mainly been used owing to their capability to capture exercise-induced metabolic responses comprehensively. Advancements in analytical technologies are expected to accelerate multi-omics integration, enabling precise characterization of physiological adaptations and supporting personalized strategies for health and performance enhancement.

Keywords: Urine; Exercise physiology; Metabolomics; Targeted; Untargeted

INTRODUCTION

Physical activity induces complex metabolic responses influenced by exercise type, intensity, duration, and individual physiological characterstics. A thorough understanding of these metabolic changes is crucial for optimizing athletic performance, improving health outcomes, and preventing disease. Metabolomics, defined as the systematic analysis of small-molecule metabolites (typically ≤1,500 Da, excluding peptides), including carbohydrates, amino acids, organic acids, nucleotides, and lip-ids, has emerged as a powerful method for studying physiological responses at the molecular level [1-3].

Human metabolism is highly complex, involving over 110,000 distinct compounds and more than 46,000 identified metabolic pathways [4]. The intricate processes in metabolism are influenced by internal and external factors, including human activities, such as physical activity, diet, lifestyle, infections, and diseases, all of which strongly affect metabolic function.

Metabolomics research in the field of exercise physiology has histori-cally centered on plasma and serum because of the comprehensive array of biomarkers present in these biofluids [5]. Invasive methods, such as muscle and tissue biopsies, have also been used extensively to investigate critical metabolic pathways, including glycolysis [6,7]. However, invasive methods are costly and operationally challenging, thereby limiting participant compliance and broader applicability [8,9]. As an alternative, urine, which is an easily accessible biofluid that is used in non-invasive methods, has gained attention in recent metabolomics research, offering practical advantages for frequent and repeated sampling [10,11].

Metabolomics research uses targeted or untargeted approaches. Targeted approaches focus on identifying and quantitatively analyzing specific metabolites, typically in predefined metabolic pathways based on hypotheses [12]. Nuclear magnetic resonance (NMR) and mass spectrometry (MS) are commonly used for analysis because of their quantitative precision and reproducibility. Advances, particularly in MS, have led to the development of targeted methods and expanded the range of metabolites that can be analyzed [13,14]. Targeted methods are also valuable for the quantitative validation of metabolites that were initially dis-covered using untargeted profiling, offering sensitive and stable analytical capabilities [15].

By comparison, untargeted metabolomics, also known as global metabolite profiling, is used for comprehensive detection of a broad spec-trum of metabolites in a sample to discover unknown metabolites and pathways, greatly reducing bias in exploring metabolic changes. Techniques include high-resolution NMR and liquid chromatography-mass spectrometry (LC-MS). LC-MS can identify an extensive range of metabolites, often numbering in the thousands per experiment [16-18]. However, untargeted methods generate large, complex datasets that need sophisticated software, such as XCMS [19] and MetAlign [20], for effi-cient data processing and analysis. Therefore, although untargeted metabolomics is suitable for exploratory research, it offers lower quantitative accuracy and reproducibility than targeted methods.

In summary, targeted metabolomics offers precise quantification for hypothesis-driven research and insights into specific metabolic pathways, whereas untargeted metabolomics provides a comprehensive metabolic profile valuable for discovering unknown metabolites and pathways with lower quantitative accuracy. Table 1 summarizes the characteristics of both approaches.

Table 1.

Characteristics of targeted and untargeted metabolomics studies

Despite the advantages and limitations of targeted and untargeted metabolomics approaches, the current body of evidence derived from urinary metabolomics studies in exercise science is fragmented. Therefore, a comprehensive synthesis of studies using both analytical strategies is essential. In the present systematic review, we sought to clarify the methodological differences, summarize current findings, identify recur-rent biomarkers, and outline potential limitations and future trends in urinary metabolomics in exercise science. We hope that our findings will provide researchers with valuable insights for monitoring physical health and improving athletic performance, thus deepening our understanding of exercise-induced metabolic adaptations.

METHODS

1. Literature search strategy

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines [21]. The literature search was performed between March 30 and April 5, 2024, and included studies published between January 1, 2004, and January 1, 2024. The search strategy used combinations of the following keywords and subject terms: “ exercise” OR “ exercise nutrition” OR “ physical activity” OR “ physical activity and nutrition” OR “ sport” OR “ sport and nutrition”, “ urine” OR “ urinary”, and “ metabolomics” OR “ metabonomics”. The following databases were used to conduct the search: Web of Science, Google Scholar, PubMed, and Scopus. The search strategy was iteratively refined by modifying the keyword combinations and the order of database queries, with repeated searches conducted on different days to ensure thorough-ness and consistency in article retrieval. Only original research articles published in English were considered for inclusion. Systematic reviews and inaccessible articles were excluded. No restrictions were placed on the study design beyond these criteria. Randomized controlled trials and feasibility studies were eligible for inclusion, and articles that included a control group were consistently included, although no direct comparisons between different interventions or populations were made. A summary of the eligibility criteria, based on the Population, Intervention, Comparison, Outcomes, and Study Design (PICOS) framework, is pre-sented in Table 2.

Table 2.

Population, Intervention, Comparator, Outcome, and Study Design (PICOS) criteria for the systematic reviews

The systematic review process included independent screening and data extraction by two reviewers, with any disagreements resolved through discussion with a third reviewer. After the initial screening, the duplicates were removed, and the remaining records were screened using title and abstract. The study selection process is illustrated in Fig. 1. In accordance with the criteria, a total of 34 studies were selected for the fi-nal analysis.

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analy-ses (PRISMA) flow diagram of article selection.

2. Selection criteria

Articles were selected if they met the following inclusion criteria: (a) published in English; (b) used urine samples as a primary biofluid for metabolomic analysis; (c) investigated metabolic responses to exercise or physical activity; (d) used either targeted or untargeted metabolomics approaches; and (e) were classified as original research articles. Studies were excluded if: (a) they were review articles, editorials, or conference ab-stracts; (b) the full text was not accessible; or (c) they did not report original metabolomic data related to exercise. No restrictions were placed on the participants’ age, sex, health status, or exercise modality. Randomized controlled trials, feasibility studies, and observational studies were included. The eligibility screening was managed using Mendeley Reference Manager 2.117.0 (Elsevier Ltd., Amsterdam, Netherlands), and a predefined coding framework was developed to code the data systematically.

3. Data items

The following data were extracted from each eligible study: (1) author(s), year of publication, and country; (2) study design and participant characteristics (e.g., sample size, sex, age, and health/training status); (3) exercise intervention details (e.g., type, intensity, duration, and fre-quency); (4) type of biological sample used and analytical platform em-ployed (e.g., NMR, LC-MS, gas chromatography [GC]-MS); (5) number and types of metabolites significantly altered; and (6) principal findings and associated metabolic pathways.

4. Collating, summarizing, and reporting the results

This phase of the review followed the methodological framework out-lined by Levac et al. [22]. A comprehensive descriptive synthesis was conducted independently by two reviewers who used a standardized charting table. The studies included were grouped and summarized according to the analytical approach (targeted vs. untargeted), platform used, and characteristics of the exercise intervention. Descriptive content analysis was then applied to compare the findings across the studies and to identify commonly reported metabolic responses. These findings were used to contextualize the current state of urinary metabolomics in exercise science and to guide future research directions in the field.

RESULTS

This section summarizes the results of the 34 studies included in this systematic review. Table 3 summarizes the studies that used targeted metabolomics approaches, and Table 4 summarizes those that used untargeted methodologies. Both tables provide details regarding study design, participant characteristics, analytical platforms, biosample types, the number of significantly altered metabolites, and key findings. Although all studies used urine as a primary biofluid, the research protocols, analytical tools, and data interpretation methods varied. The study participants ranged from adolescents to elite athletes, and the scope of metabolic profiling depended on the platform and experimental design. The overall characteristics of the included studies are summarized in Tables 3 and 4.

Table 3.

Characteristics of the studies included that used targeted metabolomics (n=5)

Table 4.

Characteristics of the studies included that used untargeted metabolomics (n=29)

1. Targeted and untargeted approaches

Five studies (14.7%) exclusively used targeted metabolomics and 29 (85.3%) used untargeted methods. Some studies combined both approaches to improve metabolite coverage and validation. Untargeted metabolomics was more prevalent, reflecting its exploratory nature and suitability for biomarker discovery. In the targeted studies, NMR was the most frequently used platform (n=3), followed by GC-time of flight (TOF)-MS (n=1), and a combined LC-MS and NMR approach (n=1). By contrast, 24 of the untargeted studies used a single detection platform, including 11 that used NMR. These findings indicate a clear preference for untargeted strategies in exercise-related urinary metabolomics.

2. Analytical platforms

The analytical platforms were categorized as NMR, MS-based, or other techniques. Overall, 14 studies (41.2%) exclusively used NMR, and five studies integrated NMR with additional methods. MS-based techniques, including LC-MS, ultraperformance liquid chromatography (UPLC)-MS, and GC-TOF-MS, were used in 12 studies (35.3%). Five studies used a combination of NMR, MS, and alternative approaches. Two studies (5.9%) used Raman spectroscopy as part of a research series. NMR was used in nine of 13 studies published before 2016 compared with nine of 21 studies published after 2016 (six used NMR exclusively and three used NMR in combination with other platforms).

3. Sample types

Although urine was the primary sample type, seven studies used multiple biofluids to complement metabolomic profiling. The sample types included urine (n=27), plasma (n=4), feces (n=3), adipose tissue (n=1), muscle tissue (n=1), exhaled air (n=1), and exhaled breath condensate (n=1). Mixed samples (n=7) were particularly useful for capturing sys-temic metabolic changes. Exhaled air and its condensate were used in the same study. Between 2010 and 2013, only three relevant studies were published, whereas nine were published between 2014 and 2015, indicating an increase in urinary metabolomics research.

4. Intervention duration

Studies were categorized based on the intervention duration: acute (≤24 hr, n=15), short-term (≤1 week, n=9), and long-term (>1 week, n=10). Acute and short-term interventions constituted the majority, reflecting an emphasis on immediate metabolic responses to exercise. Thus, short-term designs were the dominant approach in this field. These short-duration, high-intensity exercise protocols were particularly useful for capturing transient metabolic alterations and identifying exer-cise-responsive biomarkers.

5. Exercise intensity

Several studies have demonstrated that exercise intensity modulates metabolite profiles. Acute exercise was associated with increased levels of pyruvate, lactate, ketone bodies, nucleotides, and fatty acids, whereas bile acid concentrations tended to decrease. For example, Pechlivanis and colleagues [23-25] conducted a series of sprint experiments and used untargeted NMR analysis to differentiate pre- and post-exercise metabolite patterns, finding that even 30 seconds of intense exercise can lead to notable metabolic disruptions. Neal et al. [26] reported that low-intensity training altered energy metabolism markers more than high-intensity exercise. Similarly, Mukherjee et al. [27] observed that trained cyclists had distinct urinary metabolomic profiles compared with sedentary controls following submaximal endurance exercise. Another study [28] confirmed that high-intensity interval training can induce significant but rapidly reversible metabolic changes.

6. Exercise type

Different exercise types yielded distinct urinary metabolite profiles. Resistance training [29], snowboarding [30], middle-distance running [31], and swimming [32] were associated with specific shifts in metabolite compositions. Each type of exercise activates different energy metabolism pathways, which in turn affects the types and concentrations of metabolites found in urine. Studies using advanced techniques, such as Raman spectroscopy [33] or UPLC-quadrupole TOF-MS [34], demonstrated the utility of metabolomics in capturing sport-specific adaptations. For example, Kim et al. [29] documented metabolic changes following a week of soccer training in cold weather, suggesting the need for post-exercise recovery.

7. Other influencing factors

Beyond exercise modality and intensity, several studies investigated the role of diet, hypoxia, obesity, and gut microbiota. Lou et al. [35] demonstrated that hypoxic conditions significantly elevated urinary purine metabolites. Barton et al. [36] and Zheng et al. [37] linked dietary patterns and body composition to metabolite diversity and responses. Studies on whey protein supplementation [38] and eccentric exercise [39] also revealed sex-specific and nutrient-sensitive metabolic changes. For instance, Jang et al. [39] observed sex-specific differences in metabolite responses following eccentric exercise. These results highlight the com-plexity of urinary metabolomic responses to exercise, which are influenced not only by the exercise protocol (type, intensity, duration) and analytical platform, but also by various physiological, environmental, and behavioral factors. The findings from this systematic review under-score the critical need for integrative, multi-platform, and multiomics approaches in future research to characterize exercise-induced metabolic adaptations fully and their implications for health and performance.

DISCUSSION

The application of metabolomics in the field of exercise science has expanded considerably, and the focus on non-invasive biological samples, such as urine, has also increased. This systematic review synthe-sized current findings on targeted and untargeted urinary metabolomics approaches, highlighting methodological trends and the evolving role of this tool in exercise science.

Urine, which is a biofluid that can be obtained readily and non-inva-sively, offers a promising matrix for metabolomics research in exercise science. The advantages of urine over traditional biofluids, such as plasma or serum, include its suitability for repeated sampling, which is essential for monitoring dynamic metabolic changes with minimal participant burden [40,41]. Although urine was less frequently used than plasma [42], urine metabolomics avoids the invasiveness and logistical challenges associated with biospecimens, such as blood or tissue biopsies [34,43].

The rapid advancement of analytical technologies has contributed to the growth of exercise metabolomics research [44]. Despite these advancements, certain methodological limitations persist. A large number of studies continue to rely on a single analytical platform and/or use modest sample sizes, practices that can constrain metabolite detection. For instance, Prado et al. [45] used UPLC-MS^E for untargeted metabolomics analysis of urine samples collected from 30 athletes after football matches, detecting a total of 1,091 metabolites, of which 526 showed significant alterations. These included elevated levels of fatty acyls, carbox-ylic acids and their derivatives, and steroid-related metabolites. To increase metabolite detection coverage and improve quantification accuracy, recent studies have increasingly integrated multiple analytical platforms [46,47].

Variations in exercise intensity, for instance, between moderate and high-intensity physical training regimens, produce discernible metabolic responses in urine. High-intensity exercise is generally linked to the re-lease of stress-related metabolites and anaerobic energy products. By contrast, moderate-intensity physical activity results in more subtle alterations in metabolic profiles [48]. In an early application of urinary metabolomics in exercise science, Enea et al. [49] analyzed urine samples before and after prolonged and short-term high-intensity exercise in trained and untrained women. They analyzed metabolites including creatinine, lactate, pyruvate, alanine, β-hydroxybutyrate, acetate, and hypoxanthine to ascertain the effect of high-intensity exercise on urinary metabolomics. High-intensity exercise elicited significant changes in metabolites, emphasizing the sensitivity of urinary metabolomics to exercise intensity.

Urine is a convenient and widely used biological matrix for detecting a variety of drugs and their metabolites, supporting its application in clinical pharmacotherapy and drug development [50-54]. Beyond its clinical relevance, the advancement and broad adoption of urine metabolomics present great opportunities in exercise science, enabling comprehensive, multi-pathway, high-throughput investigations requiring minimally invasive or non-invasive sampling. As the principal route for metabolite excretion, urine provides a unique window into the body's dynamic metabolic state; thus, urinary metabolomics is a promising tool for elucidating exercise-induced metabolic adaptations.

Despite limitations of the current metabolomics methods used in exercise science, it is likely that these methods will become the standard for research in the field. In future, metabolomics is expected to expand into applied domains, such as sports nutrition and performance optimization. Future research may contribute to the development of personalized training and recovery strategies. The potential implications of this approach should be acknowledged, and methodologies that facilitate the integration of metabolomics with other omics techniques, such as proteomics and genomics, should be developed. Urinary metabolomics will allow researchers to capture the multifaceted biological effects of exercise comprehensively, thereby improving our understanding of the relation-ship between metabolism and exercise. Although platforms such as the Urine Metabolome Database and Metabolomics Workbench have im-proved data accessibility and reproducibility, the limited volume of exer-cise-specific data underscores the importance of establishing a globally coordinated database tailored to exercise science. Such an initiative would address inconsistencies across laboratories and facilitate international collaboration and data harmonization. Although a formal quality assessment was not performed due to the methodological heterogeneity of the included studies, potential sources of bias and design limitations were qualitatively discussed throughout this review. Furthermore, while untargeted metabolomics currently dominates the field due to its broad discovery capabilities, the translational potential of targeted approaches, particularly for biomarker validation and guiding individualized exercise prescription-should be further explored in future research.

CONCLUSION

This systematic review highlighted the growing importance of urinary metabolomics in the field of exercise science. Our descriptive synthesis reveals that untargeted metabolomics currently dominates this field, primarily because of its strong exploratory capacity for identifying novel biomarkers and metabolic pathways. Future advancements in this discipline depend on the integration of multi-platform analytical tools and multiomics strategies. To increase reproducibility, comparability, and translational value, it is essential to develop standardized research protocols and establish global, open-access metabolomics databases. Urinary metabolomics, as a minimally invasive, scalable, and informative approach, offers strong potential for elucidating exercise-induced metabolic adaptations and guiding personalized training and recovery programs. potential to improve our understanding of exercise-induced metabolic alterations and to contribute to the development of individualized training and recovery programs.

Notes

ACKNOWLEDGMENT

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manu-facturer, is not guaranteed, or endorsed by the publisher.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

Conceptualization: HS Lee, ZX Yu; Data curation: ZX Yu, JS Chang, AR Kim; Formal analysis: JS Chang; Methodology: ZX Yu, JS Chang, AR Kim; Project administration: HS Lee; Visualization: AR Kim; Writing - original draft: ZX Yu; Writing - review & editing: JS Chang, HS Lee.

References

1. . Nagana Gowda GA, Zhang S, Gu H, Asiago V, Shanaiah N, et al. Metabolomics-based methods for early disease diagnostics: a review. Ex-pert Rev Mol Diagn 2008;8:617–33.

2. . Emwas A-HM, Salek RM, Griffin JL, Merzaban J. NMR-based metabolomics in human disease diagnosis: applications, limitations, and rec-ommendations. Metabolomics 2013;9:1048–72.

3. . Keen B, Cawley A, Reedy B, Fu S. Metabolomics in clinical and forensic toxicology, sports antidoping and veterinary residues. Drug Test Anal 2022;14:794–807.

4. . Wishart DS, Li C, Marcu A, Badran H, Pon A, et al. PathBank: a comprehensive pathway database for model organisms. Nucleic Acids Res 2020;48:D470–8.

5. . Bongiovanni T, Pintus R, Dessì A, Noto A, Sardo S, et al. Sportomics: metabolomics applied to sports. The new revolution? Eur Rev Med Pharmacol Sci 2019;23:11011–9.

6. . Franchi MV, Wilkinson DJ, Quinlan JI, Mitchell WK, Lund JN, et al. Early structural remodeling and deuterium oxide-derived protein metabolic responses to eccentric and concentric loading in human skeletal muscle. Physiol Rep 2015;3:e12593.

7. . Burke MF, Dunbar RL, Rader DJ. Could exercise metabolomics pave the way for gymnomimetics? Sci Transl Med 2010;2:41ps35.

8. . Khoramipour K, Sandbakk Ø, Keshteli AH, Gaeini AA, Wishart DS, et al. Metabolomics in exercise and sports: a systematic review. Sports Med 2022;52:547–83.

9. . Kelly RS, Kelly MP, Kelly P. Metabolomics, physical activity, exercise and health: a review of the current evidence. Biochim Biophys Acta Mol Basis Dis 2020;1866:165936.

10. . Yamaguchi S, Suzuki K, Kanda K, Inami T, Okada J. Changes in urinary titin N-terminal fragments as a biomarker of exercise-induced muscle damage in the repeated bout effect. J Sci Med Sport 2020;23:536–40.

11. . Lee EC, Fragala MS, Kavouras SA, Queen RM, Pryor JL, et al. Biomarkers in sports and exercise: tracking health, performance, and recovery in Athletes. J Strength Cond Res 2017;31:2920–37.

12. . Dudley E, Yousef M, Wang Y, Griffiths WJ. Targeted metabolomics and mass spectrometry. Adv Protein Chem Struct Biol 2010;80:45–83.

13. . Davison JM, Cheyne GA. History of the measurement of glucose in urine: a cautionary tale. Med Hist 1974;18:194–7.

14. . Robinson AB, Robinson NE. Origins of metabolic profiling. Methods Mol Biol 2011;708:1–23.

15. . Johnson CH, Ivanisevic J, Siuzdak G. Metabolomics: beyond biomarkers and towards mechanisms. Nat Rev Mol Cell Biol 2016;17:451–9.

16. . Jonsson P, Bruce SJ, Moritz T, Trygg J, Sjöström M, et al. Extraction, interpretation and validation of information for comparing samples in metabolic LC/MS data sets. The Analyst 2005;130:701–7.

17. . Barton RH, Nicholson JK, Elliott P, Holmes E. High-throughput 1H NMR-based metabolic analysis of human serum and urine for large-scale epidemiological studies: validation study. Int J Epidemiol 2008;37:i31–40.

18. . Yanes O, Clark J, Wong DM, Patti GJ, Sánchez-Ruiz A, et al. Metabolic oxidation regulates embryonic stem cell differentiation. Nat Chem Biol 2010;6:411–7.

19. . Baran R, Kochi H, Saito N, Suematsu M, Soga T, et al. MathDAMP: a package for differential analysis of metabolite profiles. BMC Bioinfor-matics 2006;7:530.

20. . Lommen A. MetAlign: interface-driven, versatile metabolomics tool for hyphenated full-scan mass spectrometry data preprocessing. Anal Chem 2009;81:3079–86.

21. . Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.

22. . Levac D, Colquhoun H, O' Brien KK. Scoping studies: advancing the methodology. Implement Sci 2010;5:69.

23. . Pechlivanis A, Kostidis S, Saraslanidis P, Petridou A, Tsalis G, et al. (1) H NMR-based metabonomic investigation of the effect of two different exercise sessions on the metabolic fingerprint of human urine. J Proteome Res 2010;9:6405–16.

24. . Pechlivanis A, Kostidis S, Saraslanidis P, Petridou A, Tsalis G, et al. 1H NMR study on the short- and long-term impact of two training programs of sprint running on the metabolic fingerprint of human serum. J Proteome Res 2013;12:470–80.

25. . Pechlivanis A, Papaioannou KG, Tsalis G, Saraslanidis P, Mougios V, et al. Monitoring the response of the human urinary metabolome to brief maximal exercise by a combination of RP-UPLC-MS and (1)H NMR spectroscopy. J Proteome Res 2015;14:4610–22.

26. . Neal CM, Hunter AM, Brennan L, O' Sullivan A, Hamilton DL, et al. Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists. J Appl Physiol Bethesda 2013;114:461–71.

27. . Mukherjee K, Edgett BA, Burrows HW, Castro C, Griffin JL, et al. Whole blood transcriptomics and urinary metabolomics to define adaptive biochemical pathways of high-intensity exercise in 50-60 year old masters athletes. PloS One 2014;9:e92031.

28. . Kistner S, Rist MJ, Döring M, Dörr C, Neumann R, et al. An NMR-based approach to identify urinary metabolites associated with acute physical exercise and cardiorespiratory fitness in healthy humans-re-sults of the karmen study. Metabolites 2020;10:212.

29. . Kim HY, Lee JD, Lee YH, Seo SW, Lee HS, et al. Urinary metabolomics in young soccer players after winter training season. Metabolites 2022;12:1283.

30. . Wang F, Han J, He Q, Geng Z, Deng Z, et al. Applying (1)H NMR spectroscopy to detect changes in the urinary metabolite levels of Chi-nese half-pipe snowboarders after different exercises. J Anal Methods Chem 2015;2015:315217.

31. . Sun T, Wu Y, Wu X, Ma H. Metabolomic profiles investigation on athletes' urine 35 minutes after an 800-meter race. J Sports Med Phys Fitness 2017;57:839–49.

32. . Moreira LP, Silveira L, Da Silva AG, Fernandes AB, Pacheco MTT, et al. Raman spectroscopy applied to identify metabolites in urine of physically active subjects. J Photochem Photobiol B 2017;176:92–9.

33. . Moreira LP, Silveira L, Pacheco MTT, Da Silva AG, Rocco DDFM. Detecting urine metabolites related to training performance in swimming athletes by means of Raman spectroscopy and principal component analysis. J Photochem Photobiol B 2018;185:223–34.

34. . Zhao J, Wang Y, Zhao D, Zhang L, Chen P, et al. Integration of metabolomics and proteomics to reveal the metabolic characteristics of high-intensity interval training. The Analyst 2020;145:6500–10.

35. . Lou BS, Wu PS, Liu Y, Wang JS. Effects of acute systematic hypoxia on human urinary metabolites using LC-MS-based metabolomics. High Alt Med Biol 2014;15:192–202.

36. . Barton W, Penney NC, Cronin O, Garcia-Perez I, Molloy MG, et al. The microbiome of professional athletes differs from that of more sedentary subjects in composition and particularly at the functional metabolic level. Gut 2018;67:625–33.

37. . Zheng H, Yde CC, Arnberg K, Mølgaard C, Michaelsen KF, et al. NMR-based metabolomic profiling of overweight adolescents: an elucidation of the effects of inter-/intraindividual differences, gender, and pubertal development. BioMed Res Int 2014;2014:537157.

38. . Cronin O, Barton W, Skuse P, Penney NC, Garcia-Perez I, et al. A pro-spective metagenomic and metabolomic analysis of the impact of exercise and/or whey protein supplementation on the gut microbiome of sedentary adults. mSystems 2018;3:e00044–18.

39. . Jang HJ, Lee JD, Jeon HS, Kim AR, Kim S, et al. Metabolic profiling of eccentric exercise-induced muscle damage in human urine. Toxicol Res 2018;34:199–210.

40. . Dunn WB, Broadhurst DI, Atherton HJ, Goodacre R, Griffin JL. Sys-tems level studies of mammalian metabolomes: the roles of mass spectrometry and nuclear magnetic resonance spectroscopy. Chem Soc Rev 2011;40:387–426.

41. . Bouatra S, Aziat F, Mandal R, Guo AC, Wilson MR, et al. The human urine metabolome. Dzeja P, editor. PLoS ONE 2013;8:e73076.

42. . Bathen TF, Jensen LR, Sitter B, Fjösne HE, Halgunset J, et al. MR-deter-mined metabolic phenotype of breast cancer in prediction of lymphatic spread, grade, and hormone status. Breast Cancer Res Treat 2007;104:181–9.

43. . San-Millán I. Blood biomarkers in sports medicine and performance and the future of metabolomics. Methods Mol Biol Clifton NJ 2019;1978:431–46.

44. . Resende NM, de Magalhães Neto AM, Bachini F, de Castro LEV, Bassi-ni A, et al. Metabolic changes during a field experiment in a world-class windsurfing athlete: a trial with multivariate analyses. Omics J Integr Biol 2011;15:695–704.

45. . Prado E, Souza GHMF, Pegurier M, Vieira C, Lima-Neto ABM, et al. Non-targeted sportomics analyses by mass spectrometry to under-stand exercise-induced metabolic stress in soccer players. Int J Mass Spectrom 2017;418:1–5.

46. . Jaguri A, Al Thani AA, Elrayess MA. Exercise metabolome: insights for health and performance. Metabolites 2023;13:694.

47. . Muniz-Santos R, Magno-França A, Jurisica I, Cameron LC. From mi-crocosm to macrocosm: the -omics, multiomics, and sportomics approaches in exercise and sports. Omics J Integr Biol 2023;27:499–518.

48. . Hargreaves M, Spriet LL. Skeletal muscle energy metabolism during exercise. Nat Metab 2020;2:817–28.

49. . Enea C, Seguin F, Petitpas-Mulliez J, Boildieu N, Boisseau N, et al. (1) H NMR-based metabolomics approach for exploring urinary metabolome modifications after acute and chronic physical exercise. Anal Bioanal Chem 2010;396:1167–76.

50. . Narduzzi L, Dervilly G, Audran M, Le Bizec B, Buisson C. A role for metabolomics in the antidoping toolbox? Drug Test Anal 2020;12:677–90.

51. . Ishii A, Sato K, Kusakabe K, Kato N, Wada T. Identification and quantitative analysis of 2-fluoromethamphetamine and its metabolites in human urine. J Anal Toxicol 2023;47:59–65.

52. . Li B, He X, Jia W, Li H. Novel applications of metabolomics in personalized medicine: a mini-review. Mol Basel Switz 2017;22:1173.

53. . Jacob M, Lopata AL, Dasouki M, Abdel Rahman AM. Metabolomics toward personalized medicine. Mass Spectrom Rev 2019;38:221–38.

54. . Szeremeta M, Pietrowska K, Niemcunowicz-Janica A, Kretowski A, Ciborowski M. Applications of metabolomics in forensic toxicology and forensic medicine. Int J Mol Sci 2021;22:3010.

55. . Sheedy JR, Gooley PR, Nahid A, Tull DL, McConville MJ, et al. ¹ H-NMR analysis of the human urinary metabolome in response to an 18-month multi-component exercise program and calcium-vitamin-D₃ supplementation in older men. Appl Physiol Nutr Metab 2014;39:1294–304.

56. . Meucci M, Baldari C, Guidetti L, Alley JR, Cook C, et al. Metabolomic shifts following play-based activity in overweight preadolescents. Curr Pediatr Rev 2017;13:144–51.

57. . Kistner S, Rist MJ, Krüger R, Döring M, Schlechtweg S, et al. High-in-tensity interval training decreases resting urinary hypoxanthine con-centration in young active men-a metabolomic approach. Metabolites 2019;9:137.

58. . Pechlivanis A, Kostidis S, Saraslanidis P, Petridou A, Tsalis G, et al. (1) H NMR-based metabonomic investigation of the effect of two different exercise sessions on the metabolic fingerprint of human urine. J Proteome Res 2010;9:6405–16.

59. . Krug S, Kastenmüller G, Stückler F, Rist MJ, Skurk T, et al. The dynamic range of the human metabolome revealed by challenges. FASEB J Off Publ Fed Am Soc Exp Biol 2012;26:2607–19.

60. . Ma H, Liu X, Wu Y, Zhang N. The intervention effects of acupuncture on fatigue induced by exhaustive physical exercises: a metabolomics investigation. Evid Based Complement Alternat Med 2015;2015:1–11.

61. . Glynn EL, Piner LW, Huffman KM, Slentz CA, Elliot-Penry L, et al. Im-pact of combined resistance and aerobic exercise training on branched-chain amino acid turnover, glycine metabolism and insulin sensitivity in overweight humans. Diabetologia 2015;58:2324–35.

62. . Pechlivanis A, Papaioannou KG, Tsalis G, Saraslanidis P, Mougios V, et al. Monitoring the response of the human urinary metabolome to brief maximal exercise by a combination of RP-UPLC-MS and¹ H NMR spectroscopy. J Proteome Res 2015;14:4610–22.

63. . Daskalaki E, Blackburn G, Kalna G, Zhang T, Anthony N, et al. A study of the effects of exercise on the urinary metabolome using nor-malisation to individual metabolic output. Metabolites 2015;5:119–39.

64. . Muhsen Ali A, Burleigh M, Daskalaki E, Zhang T, Easton C, et al. Metabolomic profiling of submaximal exercise at a standardised relative intensity in healthy adults. Metabolites 2016;6:9.

65. . Jing L, Chengji W. GC/MS-based metabolomics strategy to analyze the effect of exercise intervention in diabetic rats. Endocr Connect 2019;8:654–60.

66. . Quintas G, Reche X, Sanjuan-Herráez JD, Martínez H, Herrero M, et al. Urine metabolomic analysis for monitoring internal load in professional football players. Metabolomics Off J Metabolomic Soc 2020;16:45.

67. . Cao B, Liu S, Yang L, Chi A. Changes of differential urinary metabolites after high-intensive training in teenage football players. BioMed Res Int 2020;2020:2073803.

68. . Pintus R, Bongiovanni T, Corbu S, Francavilla VC, DessÌ A, et al. Sportomics in professional soccer players: metabolomics results during pre-season. J Sports Med Phys Fitness 2021;61:324–30.

69. . Babu AF, Csader S, Männistö V, Tauriainen M-M, Pentikäinen H, et al. Effects of exercise on NAFLD using non-targeted metabolomics in adipose tissue, plasma, urine, and stool. Sci Rep 2022;12:6485.

70. . Muli S, Brachem C, Alexy U, Schmid M, Oluwagbemigun K, et al. Ex-ploring the association of physical activity with the plasma and urine metabolome in adolescents and young adults. Nutr Metab 2023;20:23.

71. . Glass KA, Germain A, Huang YV, Hanson MR. Urine metabolomics exposes anomalous recovery after maximal exertion in female ME/CFS Patients. Int J Mol Sci 2023;24:3685.

72. . Giampaoli O, Ieno C, Sciubba F, Spagnoli M, Miccheli A, et al. Metabolic biomarkers of red beetroot juice intake at rest and after physical exercise. Nutrients 2023;15:2026.

Article information Continued

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

	Targeted	Untargeted
Research approach	Analyzes specific or predefined metabolites, typically used to validate specific hypotheses	Analyzes all metabolites in a sample, suitable for discovering unknown metabolites and metabolic pathways
Sample analysis	Usually involves quantitative analysis	Usually involves semi-quantitative analysis
Metabolites analysis	Focuses on a limited number of metabolites and specific metabolic pathways; offers high precision but low coverage	Analyzes typically all metabolites in the sample; provides low precision but broad coverage
Data processing	Involves low data volume and accurate processing of known metabolites	Generates a large amount of data, requiring extensive data preprocessing and statistical or multivariate analysis

PICOS Item	Criteria
Population (P)	Human participants and animal models (e.g., rats) engaged in physical activity or exercise, regardless of age, sex, health status, or strain
Intervention (I)	Exercise or sports interventions, including aerobic, anaerobic, resistance, or mixed protocols analyzed through urinary metabolomics
Comparison (C)	Not applicable. While control groups were included when present, direct comparisons across interventions or populations were not required
Outcome Measures (O)	Identification of metabolic changes or biomarkers through urinary metabolomics analysis, using targeted or untargeted approaches
Study Design (S)	Original research articles including randomized controlled trials, feasibility studies, and other observational designs

Study Design	Subjects	Study protocol	Analytical Platform	Biosample	Number of Significantly Altered Metabolites	Key Findings
Enea et al. 2010 [49] Repeated Measures Design	22 trained and untrained females	Prolonged exercise at 75%VO_2max until exhaustion vs. short-term intensive exercise	NMR	Urine	7	Urinary metabolism changes with training status after short-term intensive exercise testing.
Sheedy et al. 2014 [55] RCT	80 individuals	Resistance training 3 times per week for 18 months	NMR	Urine	14	Significant alterations were observed in creatinine, choline, guanosine, hypoxanthine, and other metabolite levels after exercise intervention.
Meucci et al. 2017 [56] RCT	22 active overweight preadolescents	4 or 8 weeks of supervised play-based physical program, 6hr per day, 5 times per week	GC-TOF-MS	Urine	7	After 8 weeks of PA intervention, significant metabolomic differences were observed between the pre- and post-exercise samples, with no significant changes in the 4 weeks.
Kistner et al. 2019 [57] RCT	20 healthy individuals	Pedaling for 5 min, subsequently by 15 watts every 30sec until exhaustion	LC-MS, NMR	Urine	7	Metabolism rapidly recovered from acute disturbances caused by exhaustive HIIT.
Kistner et al. 2020 [28] Cross-Sectional Study	255 healthy individuals	Pedaling at 25 watts, then increase by 25 watts every 2 min until exhaustion	NMR	Urine	37	Metabolites partially reflect metabolic changes associated with acute exercise but do not provide conclusive information about individual fitness levels.

Table 4.

Characteristics of the studies included that used untargeted metabolomics (n=29)

Study Design	Subjects	Study protocol	Analytical solutions	Biosample	Number of Significantly Altered Metabolites	Result
Pechlivanis et al. 2010 [58] Repeated measures design	12 healthy males	3 sets of two 80 m runs at maximum effort	NMR^*	Urine	22	Pre- and post-groups can be distinguished by urinary metabolomics profiles, with shorter intervals between groups resulting in greater changes in the metabolomic profiles.
Krug et al. 2012 [59] Non-randomized crossover trial	15 healthy males	30 min of anaerobic intensity pedaling on a cycle ergometer	MS/MS, NMR^*	Plasma, Urine, Exhaled air, Breath condensate	11	Following the challenge onset, metabolic profiles immediately change but return to baseline within 2hr post-challenge.
Neal et al. 2013 [26] Randomized crossover design	11 male cyclists	6 weeks cycling training	NMR^*	Urine	5	Some cellular energy stress markers changed with low-intensity but not high-intensity training.
Mukherjee et al. 2014 [27] Cross-sectional study	9 competitive cyclists and 8 untrained male controls	Acute bout of submaximal endurance exercise	NMR^*	Urine	8	Post-exercise, the metabolic profiles of well-trained athletes differ from those of sedentary elderly individuals.
Zheng et al. 2014 [37] Cohort study	192 overweight adolescents aged 12 to 15 years old	Pedometer workouts for 7 consecutive days	NMR^*	Urine, plasma	5	No strong correlation was found between daily exercise and metabolomics in both plasma and urine.
Lou et al. 2014 [35] Randomized crossover study	6 sedentary healthy males.	Sitting for two hours under nor-moxia (21% O₂), hypoxia (12% O₂, ≈4,500 m altitude), and 15% O₂ (≈3,000 m altitude)	LC-TOF-MS^*	Urine	26	Purine metabolites, including uric acid, xanthine, and hypoxanthine, rise with hypoxia, serving as hypoxic markers.
Wang et al. 2015 [30] Repeated measures design	12 professionals snowboarders	Strength, endurance, and trampoline exercises	NMR^*	Urine	12	After prolonged training, organisms achieve a relatively stable physiological state to adjust to the training load.
Ma et al. 2015 [60] RCT	14 moderately trained male	Sixteen 500 m running at maximum speed	NMR^*	Urine	21	Acupuncture treatment reduced fatigue by regulating disrupted energy and choline metabolism, and quickly alleviating reactive oxygen species stress.
Glynn et al. 2015 [61] Non-RCT	19 overweight or normal individuals	Running at 65-80% of VO_2max and resistance training 3-4 times a week	LC-MS/MS^*, GC/MS	Urine and muscle tissue	5	A mechanism has been identified wherein excess acyl groups derived from BCAA and aromatic amino acid metabolism are efficiently eliminated via glycine conjugation in the liver.
Pechlivanis et al. 2015 [62] Repeated measures design	17 healthy male	3 sets of two 80 m runs. The first and second groups are 10 min apart; the second and	RP-UPLC-MS, NMR^*	Urine	26	Even 30 sec of maximal exercise induces significant metabolic disruptions, with some effects persisting for at least 2 hr.
Daskalaki et al. 2015 [63] Quasi-experimental study	3 healthy non-smoking male	third, 10 sec 1 hr treadmill workout	LC-MS^*	Urine	59	Urocortisol glucuronide exhibited a significant increase in the initial post-exercise sample.
Muhsen Ali et al. 2016 [64] Quasi-experimental study	10 recreationally active individuals	Pedalling for 45 min	LC-MS^*	Urine	19	Principal component analysis can discriminate between pre- and post-exercise samples.
Moreira et al. 2017 [32] Non-randomized crossover trial	9 physically active individuals	30 min pedaling or running	Raman spectroscopy^*	Urine	7	After 30 min of aerobic exercise, concentrations of metabolites such as urea, creatinine, ketone bodies, nitrogenous compounds, and phosphates varied.
Sun et al. 2017 [31] Non-randomized crossover trial	19 male athletes	800-meter race	NMR^*	Urine	16	Post-race changes in urinary metabolomics indicate increased oxidative stress.
Prado et al. 2017 [45] Observational study	30 male semi-professional soccer players	Soccer match	UPLC-MS^E^*	Urine	526	Hypoxanthine and associated metabolites exhibited up-regulation in urine following the soccer match, indicating increased AMP deamination.
Moreira et al. 2018 [33] Observational study	23 professional swimmers and 15 university students	6 km swimming	Raman spectroscopy^*	Urine	7	Post-training swimmers exhibited higher levels of ketone bodies, creatine, and nitrogen compounds compared to both their pre-training levels and those of sedentary subjects.
Jang et al. 2018 [39] Non-RCT	5 male and 6 female	30 min Bench-stepping	NMR^*	Urine	11	Some urinary metabolites significantly increased after eccentric exercise.
Barton et al. 2018 [36] Cross-sectional study	40 male international rugby union players and 46 controls	Rigorous training camp	NMR^*, HILIC, RP UPLC-MS, GC-MS	Urine and feces	20	Significant metabolomic differences were observed between athletes and controls in urine, with smaller differences in fecal samples.
Cronin et al. 2018 [38] RCT	90 healthy individuals	RPE rating of 5-7 for aerobics and one set of eight for three sets of 70% 1 RM resistance exercise for 8 weeks	RP, HILIC UPLC-MS and NMR^*	Urine and feces	4	Consistent consumption of whey protein induces notable changes in the gut virome composition.
Jing et al. 2019 [65] Interventional study in animals	30 type 2 diabetic rats	Five consecutive days of treadmill exercise, with each session lasting 60 min	GC/MS^*	Urine	13	Exercise exhibited anti-hyperglycemic and anti-hyperlipidemic effects and had antioxidant properties that partially prevented complications associated with diabetes mellitus.
Quintas et al. 2020 [66] Cohort study	80 male professional football players	Entire Football League Season	UPLC-TOF-MS/MS^*	Urine	21	A significant correlation existed between external load and urinary metabolic profile, indicating biochemical pathway changes linked to long-term training adaptation.
Cao et al. 2020 [67] Quasi-experimental study	12 male teenage football players	3 sets of aerobic exercise for 6 min at 55-60 RPM	GC-TOF-MS^*	Urine	25	Exercise-induced fatigue is linked to disturbances in amino acid and energy metabolism.
Zhao et al. 2020 [34] Cohort study	23 professional football players	Yo-Yo Intermittent Recovery test	UPLC-QTOF-MS^*	Urine	59	The YYIR test resulted in significant changes in 59 urinary metabolite levels in the athletes.
Pintus et al. 2021 [68] Cohort study	21 professional football players	Preseason preparation training	NMR^*	Urine	8	Changes in urinary metabolites during observation may be associated with dietary, training, and microbiota composition.
Kim et al. 2022 [29] Cohort study	14 male teenage football players	Comprehensive training five days a week	NMR^*	Urine	75	Depending on the excessive levels of ammonia, adenine and lactic acid in the urine, a minimum of 5 days recovery time is required after training.
Ambrin et al. 2022 [69] RCT	46 NAFLD participants: intervention group (n=21), control group (n=25)	12-week high-intensity interval training	LC-MS^*	Adipose Tissue Plasma Urine Stool	-	HIIT induces metabolite changes, increasing amino acids and derivatives in adipose tissue and plasma while decreasing them in urine and feces.
Muli et al. 2023 [70] Cohort study	Subjects with full-term singleton births and birthweight over 2,500g.(Urine: n=213 Plasma: n=365)	Leisure time physical activity	UPLC-MS/MS^*	Urine and plasma	82	No associations were found between physical activity and individual metabolites in plasma or urine, or with metabolite patterns in urine, in either males or females.
Katherine et al. 2023 [71] Cross-Sectional study	8 healthy sedentary females and 10 female ME/CFS patients	Maximal cardiopulmonary exercise test	LC-MS/MS^*	Urine	15	Significant differences were found between controls and ME/CFS patients in lipid and amino acid subpathways. ME/CFS patients showed no urine metabolome changes after CPET.
Giampaoli et al. 2023 [72] Randomized Crossover Design	7 healthy individuals	30 min cycle ergometer	NMR^*	Urine	4	Exercise increased dopamine-3-O-sulfate excretion 120 min after red beetroot juice intake, suggesting it may alter absorption through different transport mechanisms.

, Untargeted; AMP, adenosine monophosphate; CPET, cardiopulmonary exercise test; GC-TOF-MS, gas chromatography-time-of-flight-mass spectrometry; BCAA, branched-chain amino acid; HILIC, hydrophilic interaction liquid chromatography; LC-MS, liquid chromatography-mass spectrometry; LC-TOF-MS, liquid chromatography-time-of-flight-mass spectrometry; LTPA, leisure time physical activity; ME/CFS, myalgic encephalomyelitis/chronic fatigue syndrome; MS/MS: tandem mass spectrometry; NMR, nuclear magnetic resonance; RCT, randomized controlled trial; RPE, rating of per-ceived exertion; RM, repetition maximum; RP-UPLC-MS, reversed-phase-ultraperformance liquid chromatography-mass spectrometry; Sig., significant; UPLC-MS, ultraperformance liquid chromatog-raphy-mass spectrometry; UPLC-MS^E, ultraperformance liquid chromatography-mass spectrometry with multi-stage acquisition, referring to Waters’ proprietary data-independent acquisition (DIA) method; UPLC-QTOF-MS, ultraperformance liquid chromatography-quadrupole time-of-flight-mass spectrometry; VO_2max, maximal aerobic capacity; YYIR, Yo-Yo intermittent recovery.