You are here

Article Text

Article menu

Download PDFPDF

XML
Cardiovascular and metabolic risk
Associations between physical activity and trimethylamine N-oxide in those at risk of type 2 diabetes
  1. Stavroula Argyridou1,2,
  2. Dennis Bernieh2,3,
  3. Joseph Henson1,2,
  4. Charlotte L Edwardson1,2,
  5. Melanie J Davies1,2,
  6. Kamlesh Khunti1,4,
  7. Toru Suzuki2,3,
  8. Thomas Yates1,2
  1. 1Diabetes Research Centre, University of Leicester, Leicester, UK
  2. 2NIHR Leicester Biomedical Research Centre, University of Leicester, Leicester, UK
  3. 3Department of Cardiovascular Sciences, University of Leicester, Leicester, UK
  4. 4NIHR Applied Research Collaborations East Midlands, University of Leicester, Leicester, UK
  1. Correspondence to Dr Thomas Yates; ty20{at}leicester.ac.uk

Abstract

Introduction Trimethylamine N-oxide (TMAO) has been identified as a novel gut-derived molecule that is associated with the risk of cardiometabolic diseases. However, the relationship between TMAO and physical activity is not well understood. This study prospectively investigates the association between TMAO and objectively assessed physical activity in a population at high risk of type 2 diabetes mellitus.

Research design and methods Baseline and 12-month follow-up data were used from the Walking Away from Type 2 Diabetes trial, which recruited adults at high risk of type 2 diabetes from primary care in 2009–2010. TMAO was analyzed using targeted mass spectrometry. Generalized estimating equation models with an exchangeable correlation structure were used to investigate the associations between accelerometer-assessed exposures (sedentary time, light physical activity, moderate to vigorous physical activity (MVPA)) and TMAO, adjusting for demographic, clinical and lifestyle factors in varying degrees.

Results Overall, 483 individuals had plasma samples available for the analysis of TMAO (316 (65.4%) men, 167 (34.6%) women), contributing 886 observations to the analysis. MVPA (min/day) was associated with TMAO in all models. In the fully adjusted model, each 30 min or SD difference in MVPA was associated with 0.584 μmol/L (0.070, 1.098) and 0.456 μmol/L (0.054, 0.858) lower TMAO, respectively. Sedentary time and light physical activity were not associated with TMAO in any model.

Conclusions Engagement with MVPA was associated with lower TMAO levels, suggesting a possible new mechanism underlining the inverse relationship between physical activity and cardiometabolic health.

  • physical activity
  • cardiovascular disease(s)
  • type 2 diabetes
  • diet
http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjdrc-2020-001359

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known about this subject?

    • Trimethylamine N-oxide (TMAO) is associated with the risk of cardiometabolic diseases.

    • It is unknown whether physical activity, which is fundamental to cardiometabolic health, is also associated with TMAO.

    What are the new findings?

    • Moderate to vigorous physical activity is inversely associated with TMAO in individuals at risk of type 2 diabetes.

    • This association was independent of other cardiometabolic risk factors and diet.

    How might these results change the focus of research or clinical practice?

    • Physical inactivity may be a risk factor for elevated TMAO levels.

    Introduction

    Physical activity and diet induce a myriad of physiological adaptations that can be beneficial to human cardiometabolic health, either directly or indirectly. Recent evidence has highlighted the role of the gut microbiota in cardiometabolic health through several mechanisms, such as increased production of the microbial metabolite trimethylamine N-oxide (TMAO). TMAO is a small, organic dietary compound that is generated by the gut following ingestion of dietary L-carnitine and phosphatidylcholine-rich foods such as red meat and eggs.1 TMAO generation is dependent on the gut microbiota, which first metabolizes dietary choline to trimethylamine which is then converted to TMAO in the liver by enzymes of the flavin mono-oxygenase family.1

    TMAO has been identified through untargeted metabolomic studies as a molecule that is present in systemic circulation and associated with the risk of atherosclerosis and cardiovascular disease development.2 3 Recent systematic reviews and meta-analyses have confirmed a positive dose-dependent association between TMAO plasma levels and increased cardiovascular disease risk and mortality.4–6 Several studies have also found that plasma TMAO levels are higher in people with type 2 diabetes compared with those without,7–9 although the direction of causality has been questioned.10

    Numerous studies have also assessed the relationship between diet, gut microbiome and TMAO levels,11–13 with studies observing that TMAO is increased in omnivores compared with vegans or vegetarians.1 Importantly, those with a less healthy gut microbiota profile, evidenced by a higher proportion of Firmicutes relative to Bacteroidetes and less bacterial diversity, produce more TMAO than those with a healthier profile in response to the ingestion of the same dietary stimulus.14

    Recent findings also suggest that physical activity can positively alter gut microbiota, resulting in favorable health outcomes.15 Having an active lifestyle may positively influence gut microbiota via several possible mechanisms including the alteration of gut bacterial composition and diversity.16 17 Indeed, the benefits of physical activity on the gut microbial communities appear to be independent of diet.18 Therefore, physical activity may promote a healthier gut microbiome profile that in turn produces less harmful bioactive metabolites including TMAO, particularly in older or obese individuals who are more likely to have an unhealthy gut microbiota profile.19 20

    TMAO levels are also influenced by other factors. For example, circulating levels of TMAO are highly dependent on renal function,21 with research showing that as estimated glomerular filtration rate (eGFR) decreases, TMAO levels rise proportionally in patients with chronic kidney disease.22 As higher physical activity has been associated with higher eGFR levels in some studies,23 24 it is possible that physical activity also affects TMAO through renal function.

    Although the interaction between physical activity, gut microbiota, and cardiovascular disease has gained recent interest,16 25 the link between physical activity and TMAO is currently not well documented. The aim of this study is to provide new evidence on the independent prospective association between TMAO and physical activity in a population at high risk of type 2 diabetes.

    Methods

    Study design and participants

    This study uses prospective observational data generated from the Walking Away from Type 2 Diabetes trial in those identified with a high risk of type 2 diabetes recruited form a primary healthcare setting. The design and results of the Walking Away from Type 2 Diabetes trial have been described in detail previously.26 27 In brief, individuals were recruited from 10 general practices in Leicestershire, UK, during 2009–2010. Individuals were recruited based on having a high risk of type 2 diabetes defined as impaired glucose tolerance (2-hour glucose ≥7.8 and <11.1 mmol/L), and/or impaired fasting glycemia (fasting glucose ≥6.1 and <7.0 mmol/L) or undiagnosed type 2 diabetes, using the Leicester Practice Risk Score.26 28 The score calculates risk based on six variables (age, sex, ethnicity, body mass index (BMI), family history of the disease, and antihypertensive drug usage). The study excluded people with established type 2 diabetes, type 2 diabetes diagnosed at baseline, those currently taking steroids, and those unable to take part in any walking activity. General practices were randomized to receive a standardized information leaflet (control) or the Walking Away intervention. The Walking Away intervention consisted of an initial 3-hour group-based structured educational program, delivered to small groups by trained educators and annual group-based refresher education sessions. The primary aim of the intervention was to increase physical activity through walking activity.26 27

    Anthropometric data

    Body mass (Tanita TBE 611; Tanita, West Drayton, UK) was measured to the nearest 0.1 kg with height measured to the nearest 0.5 cm. BMI was calculated by dividing mass to squared height (kg/m2). Arterial blood pressure was measured in the sitting position after resting for 5 min (Omron Healthcare, Henfield, UK); three measurements were obtained and the average of the last two measurements was used. Information on age, current smoking status, medication history and ethnicity was obtained by interview.

    Biochemical data

    Fasted blood samples were collected at a baseline visit and then 12 months later. Lipid profile (triglycerides, high-density lipoprotein (HDL) cholesterol and total cholesterol), creatinine and HbA1c were analyzed at the clinical laboratory within Leicester Royal Infirmary hospital according to standardized quality-controlled procedures. Creatinine was used to calculate eGFR using the Modification of Diet in Renal Disease equation which was categorized as normal kidney function (eGFR >90), mildly impaired kidney function (eGFR=60–90), and moderately impaired kidney function (eGFR <60).29 Remaining plasma samples were quantified in duplicate for TMAO levels using stable isotope dilution (D9-TMAO (98.0% purity)) followed by ultra-performance liquid chromatography-tandem mass spectrometry analysis. This was performed on a Shimadzu Nexera X2 LC-30AD coupled with a Shimadzu 8050 triple quadrupole mass spectrometer (Shimadzu, Kyoto, Japan), using optimized conditions of a previously described method with a coefficient of variation of 3.6% for measurements throughout the study.30 31

    Physical activity and dietary data

    Physical activity was measured by an accelerometer (ActiGraph GT3X, Pensacola, Florida, USA) worn on the right anterior axillary line above the hip for 7 consecutive days during waking hours. Accelerometers were initialized with a 15 s epoch, with data reintegrated into 60 s epochs for the purposes of this analysis. Freedson cut-points were used to derive intensity thresholds, with <100 counts/min estimating time spent sedentary, 100–1951 counts/min estimating light-intensity physical activity, and ≥1952 counts/min estimating time spent in moderate to vigorous intensity (moderate to vigorous physical activity (MVPA)).32 At least 4 valid days of data were required for inclusion in this analysis, with a valid day consisting of at least 600 min of wear time; non-wear time was defined as more than 60 min of continuous zero count.33 Data were processed using a commercially available software package (KineSoft V.3.3.76, KineSoft, Loughborough, UK; www.kinesoft.org).

    Diet was measured using the Dietary Instrument for Nutrition Education (DINE) food frequency questionnaire at baseline and 12 months, a brief self-report tool for diet assessment, which provides a score for total fat, unsaturated fat and fiber intake.34 Weekly servings of red meat (beef, pork or lamb) and fish were extracted to be used as covariates as they contain dietary precursors to TMAO generation.1 Red meat and fish consumption was coded as 0 if participants answered ‘None’ or ‘Less than 1 a week’; 1.5 if the answer was ‘1–2 a week’; 4 if the answer was ‘3–5 a week’; and 6 if the answer was ‘6 or more a week’.

    Data inclusion

    The Walking Away intervention was unsuccessful at promoting changes to sedentary time, light physical activity or MVPA at any time point.27 Hence, for the purposes of this study, the trial cohort was pooled and data were analyzed observationally using the combined cohort. From the 808 individuals included in the Walking Away from Type 2 Diabetes trial at baseline, 500 (62%) had stored plasma samples that enabled the analysis of TMAO at both baseline and 12 months, with the remainder providing insufficient blood volume for storage. Of those with TMAO data, 483 had concurrent accelerometer data at baseline or 12-month follow-up (see figure 1).

    Figure 1
    Figure 1

    Flow diagram of included participants.

    Statistical analysis

    All analyses were conducted using IBM SPSS Statistics (V.24.0). Included data were used to form an observational cohort in order to examine whether MVPA, light physical activity and sedentary time are associated with TMAO. A generalized estimating equations model with an exchangeable correlation structure was used to allow for repeated measurements at baseline and 12 months. Continuous TMAO data displayed a positive skewed distribution and were therefore analyzed using a gamma distribution with an identity link.

    Three main accelerometer-derived exposures were investigated: average light physical activity, average MVPA and average sedentary time. Model 1 was unadjusted, model 2 included age, sex, ethnicity, time, smoking status, randomization group (intervention/control) and accelerometer wear time. Model 3 was additionally adjusted for kidney function, HDL cholesterol, triglycerides, BMI, systolic blood pressure, HbA1c, lipid-lowering medication (statins and fibrates) and blood pressure medication (ACE inhibitors, β blockers, α blockers, calcium channel blockers, diuretics and angiotensin II receptor blockers). Model 4 was additionally adjusted for red meat and fish intake. Regression coefficients are reported as per 30 min difference and per SD difference in behavior.

    In order to assess whether reported associations were consistent across levels of glycemic control in this high risk of type 2 diabetes population, an interaction term for HbA1c as a continuous variable was added to model 3. P<0.05 was considered significant for main effects, and p<0.1 was considered significant for interactions.

    Sensitivity analysis

    A sensitivity analysis was carried out using model 3 and adjusted for sedentary time if MVPA was the exposure and vice versa. This allowed us to examine whether sedentary time or MVPA were independently associated with changes in TMAO plasma levels.

    Results

    Overall, 483 individuals were included (316 (65.4%) men and 167 (34.6%) women) contributing 886 observations to the analysis (see figure 1). The baseline and follow-up characteristics of the participants included in this study are shown in table 1. The mean (SD) age and BMI of included participants were 63.5 years (7.3) and 32.2 kg/m2 (5.5), respectively, with the majority of individuals included in this analysis being White European (437 (90.5%)). Online supplemental table S1 shows the baseline characteristics for 325 participants excluded due to insufficient stored plasma samples for TMAO analysis compared with the 483 included characteristics were largely similar across the included and excluded data sets.

    Supplemental material

    View this table:
    Table 1

    Demographics, metabolic, anthropometric, physical activity and dietary characteristics of included participants (n=483)

    Figure 2 (data shown in online supplemental table S2) reports the associations of differences in light physical activity, MVPA and sedentary time with TMAO using four different models.

    Figure 2
    Figure 2

    Association between physical activity exposures and trimethylamine N-oxide (TMAO). Model 1: unadjusted. Model 2: age, sex, ethnicity, smoking status, treatment group, wear time. Model 3: model 2+kidney function, high-density lipoprotein (HDL) cholesterol, triglycerides, body mass index (BMI), systolic blood pressure, HbA1c, lipid-lowering medication, blood pressure-lowering medication. Model 4: model 3+red meat and fish. MVPA, moderate to vigorous physical activity.

    Each 30 min or SD difference in MVPA per day was associated with 0.574 μmol/L (0.033, 1.116) and 0.449 μmol/L (0.026, 0.872) lower TMAO, respectively, in the unadjusted model. The association was largely unaffected with adjustment for clinical, sociodemographic or lifestyle (including dietary) factors, with each 30 min or SD difference in MVPA associated with 0.584 μmol/L (0.070, 1.098) and 0.456 μmol/L (0.054, 0.858) lower TMAO, respectively (figure 2; online supplemental table S2). Sedentary time and light-intensity physical activity were not associated with TMAO in any model (figure 2; online supplemental table S2).

    Glycemic control (HbA1c) did not modify the associations for MVPA (p=0.269 for interaction) with TMAO.

    Sensitivity analysis

    Results for sedentary time and MVPA were unaffected when mutually adjusted (table 2), with MVPA continuing to be associated with TMAO.

    View this table:
    Table 2

    Sensitivity analysis showing associations with TMAO (μmol/L) when MVPA and sedentary time were mutually adjusted

    Discussion

    TMAO, a novel gut-derived metabolite, has increasingly been implicated with an elevated risk of developing cardiometabolic diseases,2–9 with some, but not all, research suggesting these associations are causal.10 35 In this study, MVPA was inversely associated with TMAO plasma levels independently of age, sex, ethnicity, smoking status, kidney function, HDL cholesterol, triglycerides, BMI, systolic blood pressure, HbA1c, medication, and red meat and fish consumption. Each 30 min/day of MVPA was associated with 0.584 μmol/L lower TMAO. Sedentary time and light physical activity were not found to be associated with TMAO, suggesting intensity of movement may be important. These results highlight a potential new mechanism for the observed relationship between MVPA and cardiometabolic risk. To our knowledge, this is the first study to investigate the association between physical activity and TMAO.

    It is widely established that MVPA is associated with improvements in cardiometabolic health, with previously elucidated mechanisms including pathways related to insulin resistance, lipid metabolism, and chronic low-grade inflammation.36–38 However, traditional risk factors and mechanisms do not explain all of the association between physical activity and health outcomes.39–41 In this study, adjustment for a wide range of clinical markers that have the potential to influence TMAO, including kidney function, did not attenuate association with MVPA suggesting alternative pathways may, at least in part, mediate reported associations. Gut microbiota have been implicated as an endocrine organ that plays an important role in the regulation of the hosts’ cardiometabolic health through modulating levels of bioactive metabolites, of which TMAO is one. The amount of TMAO produced in relation to choline and betaine challenge has been associated with the profile and diversity of the gut microbiome.1 2 42 As the profile and diversity of the gut microbiome have also been associated with MVPA, particularly in older or obese adults,16 17 43 MVPA may help promote a healthier gut microbiota environment that is less conducive to TMAO generation, providing an important hypothesis for future research targeting the manipulation of the microbiome and its metabolites in the promotion of cardiometabolic health.

    Although this is the first study to specifically focus on the association between physical activity and TMAO, others have investigated the effect of lifestyle interventions that have included a physical activity component, with equivocal findings. A small experimental study found no change in TMAO following a eucaloric diet combined with exercise.44 Similarly, post hoc analysis of the Tübingen Lifestyle Intervention Program study did not observe any changes in TMAO following a lifestyle intervention (targeting diet, weight loss and physical activity) in those at risk of type 2 diabetes.45 However, when the cohort was divided into tertiles of change in TMAO, those that did decrease their TMAO following the intervention were found to have higher baseline TMAO. The average baseline TMAO (μmol/L) levels in the current study were higher than in Tübingen Lifestyle Intervention Program. Given that the effect of exercise training on the gut microbiota environment has been shown to be depended on obesity status,43 physical activity and lifestyle may be particularly important interventions for targeting TMAO in metabolically high-risk obese individuals where TMAO is already elevated. However, given the dearth of evidence in this area, further intervention studies are needed to specifically investigate the impact of physical activity and exercise training on TMAO.

    This study has important strengths and limitations. The main strength is that it provides novel evidence that TMAO may be modulated by lifestyle behaviors beyond diet. The use of objective measures of physical activity and sedentary behavior to accurately capture intensity of physical activity is also a strength. However, it is important to acknowledge some limitations. The Walking Away intervention was not successful at initiating behavior change in MVPA, light-physical activity or sedentary time, therefore the Walking Away study cohort was used to investigate observational associations with TMAO rather than the effects of behavior change. Limitations specific to observational research therefore apply, including the inability to ascribe causation and the potential for confounding by unmeasured, or residual confounding by poorly measured factors. Moreover, while the use of DINE food frequency questionnaire allows for inclusion and adjustment for the main TMAO precursors or sources (red meat and fish), it does not provide assessment of all potential precursors (ie, eggs and dairy products). Therefore, it is possible that reported results for MVPA were affected by residual or unmeasured dietary confounding. Finally, the sample was recruited on the basis of having risk factors for type 2 diabetes and may not therefore be representative of the wider population.

    In conclusion, this study suggests that more time spent in MVPA may be associated with lower TMAO, independent of other cardiometabolic risk factors and diet. However, given the observational nature of this study, further research exploring the effects of changing physical activity volume and intensity on TMAO is warranted.

    Perspective

    Physical activity is fundamental to cardiometabolic health status, with the most investigated pathways related to insulin resistance, lipid metabolism, and chronic low-grade inflammation.36–38 Gut microbiota and associated metabolites may also have a role to play.16 17 43 This study provides novel evidence that TMAO, one of the most researched gut metabolites which is hypothesized to have a deleterious effect on cardiometabolic health, is inversely associated with MVPA and that this relationship was present independent of red meat consumption in individuals at risk of type 2 diabetes mellitus. These findings suggest a potential novel mechanism underpinning the inverse relationship between physical activity and cardiometabolic disease, highlighting that physical inactivity may be a risk factor for elevated TMAO levels.

    Acknowledgments

    We thank the participants of the Walking Away study.

    References

    1. ↵
      1. Koeth RA,
      2. Wang Z,
      3. Levison BS, et al
      . Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med 2013;19:576–85.doi:10.1038/nm.3145pmid:http://www.ncbi.nlm.nih.gov/pubmed/23563705
      OpenUrlCrossRefPubMed
    2. ↵
      1. Wang Z,
      2. Klipfell E,
      3. Bennett BJ, et al
      . Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011;472:57–63.doi:10.1038/nature09922pmid:http://www.ncbi.nlm.nih.gov/pubmed/21475195
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Tang WHW,
      2. Wang Z,
      3. Levison BS, et al
      . Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk. N Engl J Med 2013;368:1575–84.doi:10.1056/NEJMoa1109400pmid:http://www.ncbi.nlm.nih.gov/pubmed/23614584
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Schiattarella GG,
      2. Sannino A,
      3. Toscano E, et al
      . Gut microbe-generated metabolite trimethylamine-N-oxide as cardiovascular risk biomarker: a systematic review and dose-response meta-analysis. Eur Heart J 2017;38:2948–56.doi:10.1093/eurheartj/ehx342pmid:http://www.ncbi.nlm.nih.gov/pubmed/29020409
      OpenUrlCrossRefPubMed
    5. ↵
      1. Qi J,
      2. You T,
      3. Li J, et al
      . Circulating trimethylamine N-oxide and the risk of cardiovascular diseases: a systematic review and meta-analysis of 11 prospective cohort studies. J Cell Mol Med 2018;22:185–94.doi:10.1111/jcmm.13307pmid:http://www.ncbi.nlm.nih.gov/pubmed/28782886
      OpenUrlCrossRefPubMed
    6. ↵
      1. Heianza Y,
      2. Ma W,
      3. Manson JE, et al
      . Gut microbiota metabolites and risk of major adverse cardiovascular disease events and death: a systematic review and meta-analysis of prospective studies. J Am Heart Assoc 2017;6. doi:doi:10.1161/JAHA.116.004947. [Epub ahead of print: 29 Jun 2017].pmid:http://www.ncbi.nlm.nih.gov/pubmed/28663251
    7. ↵
      1. Lever M,
      2. George PM,
      3. Slow S, et al
      . Betaine and trimethylamine-N-oxide as predictors of cardiovascular outcomes show different patterns in diabetes mellitus: an observational study. PLoS One 2014;9:e114969. doi:10.1371/journal.pone.0114969pmid:http://www.ncbi.nlm.nih.gov/pubmed/25493436
      OpenUrlCrossRefPubMed
    8. ↵
      1. Shan Z,
      2. Sun T,
      3. Huang H, et al
      . Association between microbiota-dependent metabolite trimethylamine-N-oxide and type 2 diabetes. Am J Clin Nutr 2017;106:888–94.doi:10.3945/ajcn.117.157107pmid:http://www.ncbi.nlm.nih.gov/pubmed/28724646
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Svingen GFT,
      2. Schartum-Hansen H,
      3. Pedersen ER, et al
      . Prospective associations of systemic and urinary choline metabolites with incident type 2 diabetes. Clin Chem 2016;62:755–65.doi:10.1373/clinchem.2015.250761pmid:http://www.ncbi.nlm.nih.gov/pubmed/26980210
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Jia J,
      2. Dou P,
      3. Gao M, et al
      . Assessment of causal direction between gut microbiota-dependent metabolites and cardiometabolic health: a bidirectional Mendelian randomization analysis. Diabetes 2019;68:1747–55.doi:10.2337/db19-0153pmid:http://www.ncbi.nlm.nih.gov/pubmed/31167879
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Aron-Wisnewsky J,
      2. Clément K
      . The gut microbiome, diet, and links to cardiometabolic and chronic disorders. Nat Rev Nephrol 2016;12:169–81.doi:10.1038/nrneph.2015.191pmid:http://www.ncbi.nlm.nih.gov/pubmed/26616538
      OpenUrlCrossRefPubMed
    12. ↵
      1. Hartiala J,
      2. Bennett BJ,
      3. Tang WHW, et al
      . Comparative genome-wide association studies in mice and humans for trimethylamine N-oxide, a proatherogenic metabolite of choline and L-carnitine. Arterioscler Thromb Vasc Biol 2014;34:1307–13.doi:10.1161/ATVBAHA.114.303252pmid:http://www.ncbi.nlm.nih.gov/pubmed/24675659
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. David LA,
      2. Maurice CF,
      3. Carmody RN, et al
      . Diet rapidly and reproducibly alters the human gut microbiome. Nature 2014;505:559–63.doi:10.1038/nature12820pmid:http://www.ncbi.nlm.nih.gov/pubmed/24336217
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Cho CE,
      2. Taesuwan S,
      3. Malysheva OV, et al
      . Trimethylamine-N-Oxide (TMAO) response to animal source foods varies among healthy young men and is influenced by their gut microbiota composition: a randomized controlled trial. Mol Nutr Food Res 2017;61. doi:doi:10.1002/mnfr.201600324. [Epub ahead of print: 03 Aug 2016].pmid:http://www.ncbi.nlm.nih.gov/pubmed/27377678
    15. ↵
      1. Fiuza-Luces C,
      2. Santos-Lozano A,
      3. Joyner M, et al
      . Exercise benefits in cardiovascular disease: beyond attenuation of traditional risk factors. Nat Rev Cardiol 2018;15:731. doi:10.1038/s41569-018-0065-1pmid:http://www.ncbi.nlm.nih.gov/pubmed/30115967
      OpenUrlPubMed
    16. ↵
      1. Chen J,
      2. Guo Y,
      3. Gui Y, et al
      . Physical exercise, gut, gut microbiota, and atherosclerotic cardiovascular diseases. Lipids Health Dis 2018;17:17. doi:10.1186/s12944-017-0653-9pmid:http://www.ncbi.nlm.nih.gov/pubmed/29357881
      OpenUrlPubMed
    17. ↵
      1. Denou E,
      2. Marcinko K,
      3. Surette MG, et al
      . High-Intensity exercise training increases the diversity and metabolic capacity of the mouse distal gut microbiota during diet-induced obesity. Am J Physiol Endocrinol Metab 2016;310:E982–93.doi:10.1152/ajpendo.00537.2015pmid:http://www.ncbi.nlm.nih.gov/pubmed/27117007
      OpenUrlCrossRefPubMed
    18. ↵
      1. Evans CC,
      2. LePard KJ,
      3. Kwak JW, et al
      . Exercise prevents weight gain and alters the gut microbiota in a mouse model of high fat diet-induced obesity. PLoS One 2014;9:e92193. doi:10.1371/journal.pone.0092193pmid:http://www.ncbi.nlm.nih.gov/pubmed/24670791
      OpenUrlCrossRefPubMed
    19. ↵
      1. Mariat D,
      2. Firmesse O,
      3. Levenez F, et al
      . The Firmicutes/Bacteroidetes ratio of the human microbiota changes with age. BMC Microbiol 2009;9:123. doi:10.1186/1471-2180-9-123pmid:http://www.ncbi.nlm.nih.gov/pubmed/19508720
      OpenUrlCrossRefPubMed
    20. ↵
      1. Ley RE,
      2. Turnbaugh PJ,
      3. Klein S, et al
      . Microbial ecology: human gut microbes associated with obesity. Nature 2006;444:1022–3.doi:10.1038/4441022apmid:http://www.ncbi.nlm.nih.gov/pubmed/17183309
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. DiNicolantonio JJ,
      2. McCarty M,
      3. OKeefe J
      . Association of moderately elevated trimethylamine N-oxide with cardiovascular risk: is TMAO serving as a marker for hepatic insulin resistance. Open Heart 2019;6:e000890. doi:10.1136/openhrt-2018-000890pmid:http://www.ncbi.nlm.nih.gov/pubmed/30997120
      OpenUrlFREE Full Text
    22. ↵
      1. Missailidis C,
      2. Hällqvist J,
      3. Qureshi AR, et al
      . Serum trimethylamine-N-oxide is strongly related to renal function and predicts outcome in chronic kidney disease. PLoS One 2016;11:e0141738. doi:10.1371/journal.pone.0141738pmid:http://www.ncbi.nlm.nih.gov/pubmed/26751065
      OpenUrlCrossRefPubMed
    23. ↵
      1. Robinson-Cohen C,
      2. Littman AJ,
      3. Duncan GE, et al
      . Physical activity and change in estimated GFR among persons with CKD. J Am Soc Nephrol 2014;25:399–406.doi:10.1681/ASN.2013040392pmid:http://www.ncbi.nlm.nih.gov/pubmed/24335971
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Hawkins MS,
      2. Sevick MA,
      3. Richardson CR, et al
      . Association between physical activity and kidney function: National health and nutrition examination survey. Med Sci Sports Exerc 2011;43:1457–64.doi:10.1249/MSS.0b013e31820c0130pmid:http://www.ncbi.nlm.nih.gov/pubmed/21200336
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. O'Sullivan O,
      2. Cronin O,
      3. Clarke SF, et al
      . Exercise and the microbiota. Gut Microbes 2015;6:131–6.doi:10.1080/19490976.2015.1011875pmid:http://www.ncbi.nlm.nih.gov/pubmed/25800089
      OpenUrlPubMed
    26. ↵
      1. Yates T,
      2. Davies MJ,
      3. Henson J, et al
      . Walking away from type 2 diabetes: trial protocol of a cluster randomised controlled trial evaluating a structured education programme in those at high risk of developing type 2 diabetes. BMC Fam Pract 2012;13:46. doi:10.1186/1471-2296-13-46pmid:http://www.ncbi.nlm.nih.gov/pubmed/22642610
      OpenUrlCrossRefPubMed
    27. ↵
      1. Yates T,
      2. Edwardson CL,
      3. Henson J, et al
      . Walking away from type 2 diabetes: a cluster randomized controlled trial. Diabet Med 2017;34:698-707.doi:10.1111/dme.13254pmid:http://www.ncbi.nlm.nih.gov/pubmed/27589017
      OpenUrlPubMed
    28. ↵
      1. Gray LJ,
      2. Taub NA,
      3. Khunti K, et al
      . The Leicester risk assessment score for detecting undiagnosed type 2 diabetes and impaired glucose regulation for use in a multiethnic UK setting. Diabet Med 2010;27:887–95.doi:10.1111/j.1464-5491.2010.03037.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/20653746
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Levey AS,
      2. Coresh J,
      3. Greene T, et al
      . Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006;145:247–54.doi:10.7326/0003-4819-145-4-200608150-00004pmid:http://www.ncbi.nlm.nih.gov/pubmed/16908915
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Suzuki T,
      2. Heaney LM,
      3. Bhandari SS, et al
      . Trimethylamine N-oxide and prognosis in acute heart failure. Heart 2016;102:841–8.doi:10.1136/heartjnl-2015-308826pmid:http://www.ncbi.nlm.nih.gov/pubmed/26869641
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Heaney LM,
      2. Jones DJL,
      3. Mbasu RJ, et al
      . High mass accuracy assay for trimethylamine N-oxide using stable-isotope dilution with liquid chromatography coupled to orthogonal acceleration time of flight mass spectrometry with multiple reaction monitoring. Anal Bioanal Chem 2016;408:797–804.doi:10.1007/s00216-015-9164-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/26573169
      OpenUrlCrossRefPubMed
    32. ↵
      1. Freedson PS,
      2. Melanson E,
      3. Sirard J
      . Calibration of the computer science and applications, Inc. accelerometer. Med Sci Sports Exerc 1998;30:777–81.doi:10.1097/00005768-199805000-00021pmid:http://www.ncbi.nlm.nih.gov/pubmed/9588623
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Evenson KR,
      2. Terry JW
      . Assessment of differing definitions of accelerometer nonwear time. Res Q Exerc Sport 2009;80:355–62.doi:10.1080/02701367.2009.10599570pmid:http://www.ncbi.nlm.nih.gov/pubmed/19650401
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Roe L,
      2. Strong C,
      3. Whiteside C, et al
      . Dietary intervention in primary care: validity of the dine method for diet assessment. Fam Pract 1994;11:375–81.doi:10.1093/fampra/11.4.375pmid:http://www.ncbi.nlm.nih.gov/pubmed/7895964
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Zhao Y,
      2. Wang Z
      . Impact of trimethylamine N-oxide (TMAO) metaorganismal pathway on cardiovascular disease. J Lab Precis Med 2020;5. doi:doi:10.21037/jlpm.2020.01.01. [Epub ahead of print: 20 Apr 2020].pmid:http://www.ncbi.nlm.nih.gov/pubmed/32587943
    36. ↵
      1. Taylor RS,
      2. Brown A,
      3. Ebrahim S, et al
      . Exercise-based rehabilitation for patients with coronary heart disease: systematic review and meta-analysis of randomized controlled trials. Am J Med 2004;116:682–92.doi:10.1016/j.amjmed.2004.01.009pmid:http://www.ncbi.nlm.nih.gov/pubmed/15121495
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Warburton DER,
      2. Nicol CW,
      3. Bredin SS
      . Health benefits of physical activity: the evidence. Can Med Assoc J 2006;174:801–9.doi:10.1503/cmaj.051351
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. Thompson PD,
      2. Buchner D,
      3. Piña IL, et al
      . Exercise and physical activity in the prevention and treatment of atherosclerotic cardiovascular disease. Circulation 2003;107:3109–16.doi:10.1161/01.CIR.0000075572.40158.77
      OpenUrlFREE Full Text
    39. ↵
      1. Reddigan JI,
      2. Ardern CI,
      3. Riddell MC, et al
      . Relation of physical activity to cardiovascular disease mortality and the influence of cardiometabolic risk factors. Am J Cardiol 2011;108:1426–31.doi:10.1016/j.amjcard.2011.07.005pmid:http://www.ncbi.nlm.nih.gov/pubmed/21855834
      OpenUrlCrossRefPubMed
    40. ↵
      1. Mora S,
      2. Cook N,
      3. Buring JE, et al
      . Physical activity and reduced risk of cardiovascular events: potential mediating mechanisms. Circulation 2007;116:2110–8.doi:10.1161/CIRCULATIONAHA.107.729939pmid:http://www.ncbi.nlm.nih.gov/pubmed/17967770
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Hamer M,
      2. Stamatakis E
      . Physical activity and risk of cardiovascular disease events: inflammatory and metabolic mechanisms. Med Sci Sports Exerc 2009;41:1206–11.doi:10.1249/MSS.0b013e3181971247pmid:http://www.ncbi.nlm.nih.gov/pubmed/19461547
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Wang Z,
      2. Tang WHW,
      3. Buffa JA, et al
      . Prognostic value of choline and betaine depends on intestinal microbiota-generated metabolite trimethylamine-N-oxide. Eur Heart J 2014;35:904–10.doi:10.1093/eurheartj/ehu002pmid:http://www.ncbi.nlm.nih.gov/pubmed/24497336
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      1. Allen JM,
      2. Mailing LJ,
      3. Niemiro GM, et al
      . Exercise alters gut microbiota composition and function in lean and obese humans. Med Sci Sports Exerc 2018;50:747–57.doi:10.1249/MSS.0000000000001495pmid:http://www.ncbi.nlm.nih.gov/pubmed/29166320
      OpenUrlCrossRefPubMed
    44. ↵
      1. Erickson ML,
      2. Malin SK,
      3. Wang Z, et al
      . Effects of lifestyle intervention on plasma trimethylamine N-oxide in obese adults. Nutrients 2019;11:179. doi:10.3390/nu11010179pmid:http://www.ncbi.nlm.nih.gov/pubmed/30654453
      OpenUrlPubMed
    45. ↵
      1. Randrianarisoa E,
      2. Lehn-Stefan A,
      3. Wang X, et al
      . Relationship of serum trimethylamine N-oxide (TMAO) levels with early atherosclerosis in humans. Sci Rep 2016;6:26745. doi:10.1038/srep26745pmid:http://www.ncbi.nlm.nih.gov/pubmed/27228955
      OpenUrlCrossRefPubMed

    Footnotes

    • Contributors SA performed statistical analysis with support from TY. SA and TY wrote the first draft of the manuscript. DB performed the TMAO assays. CLE coordinated the Walking Away study and processed accelerometer data. DB, JH, TS, MJD, KK, CLE and TY made substantial contributions to the analysis and interpretation of data, and revised the manuscript critically for intellectual content. TY is responsible for the integrity of the work as a whole. All authors gave final approval of the version to be published.

    • Funding This work was supported by the National Institute for Health Research (NIHR) Leicester Biomedical Research Centre, Leicester, UK to SA, DB, JH, TS, MJD, CLE and TY; NIHR Applied Research Collaboration East Midlands (ARC EM) to KK.

    • Disclaimer The funding sources had no role in study design, data collection, data analysis, data interpretation or writing of the report. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

    • Competing interests None declared.

    • Patient consent for publication Not required.

    • Ethics approval All participants provided written informed consent and the study was approved by the National Health Service (NHS) National Research Ethics Service-Nottingham Research Ethics Committee 2 (09/H0408/32). The study was coordinated from the Leicester Diabetes Centre, University Hospitals of Leicester NHS Trust.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement The analyzed minimal data set is available from the corresponding author (ty20@leicester.ac.uk) on reasonable request.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.