You are here

Article Text

Article menu

Download PDFPDF

Original article
Changes to the cardiac biomarkers of non-elite athletes completing the 2009 London Marathon
  1. Polly Baker1,
  2. Sarah Louise Davies2,
  3. Joanne Larkin3,
  4. Daniel Moult4,
  5. Sally Benton5,
  6. Andrew Roberts6,
  7. Timothy Harris7
  1. 1Department of Sports Medicine and Rehabilitation, Defence Medical Rehabilitation Centre, Headley Court, Surrey, UK
  2. 2Department of Sports Medicine, Rheumatology and Rehabilitation, Royal National Orthopaedic Hospital, London, Stanmore, UK
  3. 3Institute of Sports, Exercise and Health, University College London, London, UK
  4. 4Department of Anaesthesia and Intensive Care Medicine, Kings College London, London, UK
  5. 5Department Clinical Biochemistry, Royal London Hospital, Barts and the London NHS Trust, London, UK
  6. 6Department of Research, Defence Medical Rehabilitation Centre, Headley Court, Surrey, UK
  7. 7Department of Emergency Medicine and Research, Queen Marys University London, London, UK
  1. Correspondence to Dr Polly Baker, Department of Sports Medicine and Rehabilitation, Defence Medical Rehabilitation Centre, Headley Court, Headley, Epsom, Surrey KT18 6JN, UK; pollybaker{at}cantab.net

Abstract

Introduction Many studies have demonstrated a rise in troponin and brain natriuretic peptide (BNP) levels following prolonged and/or strenuous exercise. Only one study looked at athletes who collapse and this showed no difference in cardiac biomarkers between those who collapsed and those who completed without requiring medical attention. We set out to describe and quantify the changes in troponin and BNP in three groups of non-elite runners at the 2009 London marathon: those with and without known structural heart disease (SHD) and those who collapsed on completion.

Methods The first group (recruited group, RG) was recruited at the prerace exhibition. This group had two subsets, runners with SHD and without (non-SHD). A second group was recruited from those who collapsed (collapsed group, CG). Blood was taken for troponin I (TnI), troponin T (TnT), high sensitivity TnT (HSTnT) and BNP.

Results Cardiac biomarker levels increased in all groups following the marathon. No statistically significant difference was seen between the SHD and non-SHD subgroups. When comparing the RG and CG the number and degree of rise was greater in those who collapsed. A trend for the degree of rise of HSTnT was demonstrated.

Discussion We identified runners with troponin levels that, in other circumstances, would raise concern for myocardial necrosis. However absence of adverse clinical sequelae would suggest this rise is physiological. The cause and clinical significance of the increased HSTnT levels seen in those that collapsed is yet to be fully elucidated.

  • clinical assessment
  • cardiac care

https://doi.org/10.1136/emermed-2012-201465

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.

    Introduction

    Many studies have demonstrated a rise in the levels of cardiac biomarkers, including troponin, following prolonged and/or strenuous exercise.1–13 The vast majority of these studies have observed athletes without reported medical history who have completed a race without any medical event. Participation in endurance events is rising.14 ,15 With this brings an increase to the diversity of the runners’ demographic background and past medical history. There is therefore a requirement to expand the data of other runner cohorts. This will enable appropriate interpretation of cardiac biomarker results and in turn aid clinical assessment. Two such relevant groups include those with cardiovascular disease and those that collapse. Exercise-associated collapse is defined as ‘collapse in conscious athletes who are unable to stand or walk unaided after completion of an exertional event or stopping exercise’.16

    Of those runners seeking medical attention, exercise-associated collapse is the most common condition seen in the medical tent, comprising 59–85% of all visits after marathons and ultramarathons.17 ,18 Siegel et al13 is the only study to investigate cardiac biomarkers in endurance runners who have collapsed. The authors assessed the percentage of athletes with detectable troponin levels and compared these values with those reported in other studies where the runners did not collapse. They concluded that since the values were found to be very similar a cardiovascular cause to the collapse could not be assumed.

    Eijsvogels et al19 are the only researchers to investigate the effect of prolonged exercise on cardiac biomarker levels in those with cardiovascular disease. They observed the effect of a 4-day long-distance walking event (30–50 km/day) on the troponin I (TnI) levels of 109 participants of a wide age range (21–82 years). Twenty-two per cent were diagnosed with prior cardiovascular disease (including hypertension, hypercholesterolaemia and previous myocardial infarction (MI) or cerebrovascular infarction). Six per cent (6/109) showed a troponin above the level required to diagnose an acute MI. Cardiovascular disease and age in this subgroup was higher than reported for the entire group.

    We would like to add to the research in this area looking specifically at non-elite runners competing in an endurance event, in this case the London Marathon. As well as troponin we would like to study brain-natriuretic peptide (BNP). BNP is a biologically active peptide secreted in response to myocardial stretch and that opposes the activation of the renin-angiotensin-aldosterone system. Although BNP was originally used to determine severity and prognosis in heart failure, the development of rapid and inexpensive assays has allowed its role to evolve and expand. It is increasingly being used in the acute setting to differentiate between cardiac and non-cardiac causes of dyspnoea20 and it has been suggested to also play a role in identifying cardiac causes of syncope.21 Studies investigating BNP in the context of endurance events are fewer but do suggest a rise after prolonged exercise.6 ,8–11 ,13

    Although the (patho)physiological process underlying the elevated troponins seen in athletes on completion of an endurance event is not fully understood it has been suggested22 that it represents the release of unbound troponin through increased membrane permeability of myocardial cells. It is hypothesized that that this increased membrane permeability is secondary to the physiological stress placed on the cells as a result of the endurance event. On this basis it would seem logical to assume that if there were greater physiological stress placed on the myocardium, for example in the situation of a collapse secondary to hyperthermia, or if the cells were less able to cope with the physiological stress, for example in those with cardiovascular disease, the degree of rise and proportion of those with a detectable troponin would be greater. Our null hypotheses are therefore as follows:

    1. Runners would demonstrate no rise in cardiac biomarkers following an endurance event.

    2. Runners with pre-existing heart disease would have no greater increase in cardiac biomarkers than those with no pre-existing disease.

    3. Runners who collapsed would not have a greater increase in troponin than those that completed the marathon with no significant medical event.

    Methods

    Participant recruitment and group allocation

    Two study groups were defined. The first group (recruited group, RG) was recruited prerace at registration and followed up postrace on the day of the marathon. This group consisted of two subsets, runners with or without structural cardiac disease. A second group was recruited from those who collapsed towards the end of the event (collapsed group, CG).

    The participants were thus allocated to one of these groups:

    Group 1—RG

    •   1a. Those recruited premarathon with no known structural heart disease (SHD), n=57. (Non-SHD Group (NSHD)).

    •   1b. Those recruited premarathon who had a history of SHD. SHD was defined as valvular heart disease (excluding mitral valve prolapse), atrial fibrillation, ischaemic heart disease, cardiomyopathy, transplanted heart and/or hypertension with left ventricular hypertrophy, n=5. (SHD Group).

    Group 2—CG

    •   This was defined as those who were unable to stand unaided upon completing the marathon and who required emergency attention by attending medical personnel, n=14.

    Exclusion criteria

    1. Any participant with a completion time over 5 h (American College of Sports Medicine (ACSM) defines aerobic walking that is, over 5 mph and jogging/running as vigorous activity. A 5 h cut-off corresponds to this 5 mph threshold).

    2. Any participant with a completion time less than 3 h (ie, non-elite).

    3. Any chronic medical condition other than those that fit the criteria for structural heart abnormalities.

    4. Persons under 17 years.

    5. Any adult unable to comprehend the consent form.

    Data collection

    All runners completed the 26 mile London marathon. At completion of the race, recruited participants were guided to the medical area by personnel at the finish. Those recruited prerace had blood taken before and after the race. Those who collapsed only had blood taken after the marathon. Blood was taken within 15 min of completion of the event. All groups completed a questionnaire to record demographics, past medical history and medication.

    Recruited group: NSHD and SHD subgroups: blood measurement

    Three millilitres of blood was taken from the antecubital vein using the Monovette vacuum system. This was performed in the sitting position in all participants to reduce sampling errors.

    TnI and BNP were measured immediately using whole blood on the Cardio-Profiler Triage Point of Care testing device (Alere Ltd Cheshire, UK (previously Biosite)). Serum samples were simultaneously collected into Greiner Vacuette blood bottles and sent to the Royal London Hospital laboratory for troponin T (TnT) and high sensitivity TnT (HSTnT) analysis (Roche diagnostics).

    The universal definition of MI requires blood levels of troponin to be above the 99th percentile of the reference limit of the individual assay together with evidence of MI (symptoms, ECG changes). The definition requires the troponin assay to have an imprecision (coefficient of variation) at the 99th percentile of less than or equal to 10%.

    Diagnostic levels for myocardial necrosis, based on the 99th percentile of the individual troponin assay with a coefficient of variation ≤10% were 50 ng/l, 30 ng/l and 14 ng/l for TnI, TnT and HSTnT, respectively. The lower detection limits (the lowest measurable concentration that can be distinguished from zero) for the troponin assays were 50 ng/l, 10 ng/l and 3 ng/l for TnI, TnT and HSTnT, respectively. For the purpose of this study all runners with a detectable troponin result on any method were treated as positive and followed up accordingly.

    CG: blood measurement

    A blood sample was taken from every collapsed runner as part of the routine assessment—for exercise associated hyponatraemia. An additional 2 ml was taken simultaneously for our study. This was stored and only processed once the runner had recovered and had full mental capacity to consent for this.

    Before leaving the medical area, all participants were given an advice leaflet detailing symptoms that would warrant further medical attention and contact details for the study coordinators.

    Follow-up of participants

    Any participant with an elevated level of troponin before the marathon was excluded from the study and recommended to seek medical attention. Participants showing detectable levels of biomarkers after the race were contacted by telephone. They were questioned for symptoms or signs to corroborate a possible cardiac event. Those deemed at risk were recommended to seek urgent medical advice.

    Ethical review

    The study was reviewed and approved by the East London and City Alpha research ethics committees. All subjects were provided with verbal and written details of the study. Written consent was obtained.

    Statistical analyses

    Due to the data not fitting a normal distribution non-parametric tests were used to test the differences between premarathon and postmarathon cardiac biomarker levels and between different groups. Values below the level of detection were substituted for the relevant threshold levels of the corresponding assays. The Wilcoxon signed-rank test was used to compare differences between cardiac biomarkers levels premarathon and postmarathon. The following groups were compared: SHD group versus NSHD group and the RG versus CG. They were compared in terms of the degree of rise of cardiac biomarker and the proportion of those with a detectable level. The Wilcoxon-Mann-Whitney test was used with respect to the degree of rise of cardiac biomarker and the Fisher exact test was used with respect to the proportion of detectable cardiac biomarker. All statistical analyses were performed using STATA V.11.

    Results

    Hypothesis 1: Do runners that complete an endurance event demonstrate a rise in their cardiac biomarkers?

    Sixty-three runners were recruited as outlined in figure 1. One participant was immediately excluded as a result of a raised troponin level prerace. The subject was identified as having fast atrial fibrillation. He was advised not to compete and to seek medical attention. Of the remaining 62, 5 fulfilled the criteria of having SHD.

    Figure 1
    Figure 1

    Consort diagram for recruited group. BNP, brain-natriuretic peptide; NSHD, non-structural heart disease; SHD, structural heart disease TnI, troponin I; TnT, troponin T; HSTnT, high sensitivity TnT.

    Data was available for analysis in 45 (troponin) and 48 (BNP) of participants allocated to the NSHD. All data were available for the five participants with SHD.

    The levels of cardiac markers in the blood increased in all groups following the marathon. This was statistically significant (p<0.05) for all blood tests in the recruited and NSHD subgroup. Only the HSTnT biomarker showed statistical significance in the SHD subgroup. Troponin tests in all groups were below the level of detection for the assay used before the marathon. The data for this hypothesis is presented in table 1. All runners with a detectable troponin were contacted and found to be well.

    View this table:
    Table 1

    The number and proportion of runners with a detectable troponin, the means and p values for the level of troponin rise postmarathon compared with the premarathon value (significant values are bolded) are shown for non-structural heart disease (NSHD) and  structural heart disease (SHD) subgroups and the recruited group

    Hypothesis 2: Do runners with SHD have a greater increase in cardiac biomarkers than those without?

    No statistically significant difference was seen between the NSHD and SHD subgroups. This was the case for the proportion of participants with a rise in their cardiac markers (table 2) and the degree of rise of cardiac marker following the marathon (table 3).

    View this table:
    Table 2

    This table is comparing the change in the proportion of participants with a detectable level of cardiac biomarkers between non-structural heart disease (NSHD) and structural heart disease (SHD) subgroups (no participants had a detectable troponin premarathon)

    View this table:
    Table 3

    This table is comparing the change in level of cardiac biomarkers between non-structural heart disease (NSHD) and structural heart disease (SHD) subgroups premarathon and postmarathon

    Hypothesis 3: Do runners who collapse postmarathon have similar changes in troponin to those who complete without collapse?

    Sixteen collapsed runners were recruited from the medical tents at the finish into the CG (see figure 2). Two of these runners were excluded because they did not return to full mental capacity to give consent before transfer to hospital. We were unable to gain further information on these subjects from their admitting hospitals.

    Figure 2
    Figure 2

    Consort diagram for collapsed group.

    Descriptive data for the CG is detailed in table 4 (data for the RG is repeated here for ease of comparison). When comparing the RG and the CG it can be seen that the proportion of participants who developed raised troponin biomarkers was higher in those that collapsed than those that did not. This did not, however, reach statistical significance. See table 5. A greater degree of rise was also seen in the CG when compared with the RG. A trend towards higher values of the most sensitive assay (HSTnT) was seen in the CG but did not quite reach statistical significance. See table 6. We are able to compare the RG with the CG as we have already shown in our second hypothesis that there is no significant difference between the NSHD and SHD subgroups.

    View this table:
    Table 4

    The number and proportion (%) of runners with a detectable troponin and the means are shown for the collapsed and recruited group

    View this table:
    Table 5

    This table is comparing the change in the proportion of participants with a detectable level of cardiac biomarkers between the collapsed and recruited groups postmarathon

    View this table:
    Table 6

    This table is comparing the change in level of cardiac biomarkers between the collapsed and recruited groups postmarathon (the trend seen for HSTnT is underlined)

    Discussion

    The troponin levels from the groups recruited prerace were raised postmarathon. This was statistically significant (p<0.05) for the tests in the whole RG and NSHD subgroup, but only for the HSTnT in the SHD subgroup. The lack of statistical significance for TnI and TnT in the SHD group may be related to the small sample size of this group (n=5) (type 2 error).

    Using the assays described 5 out of 53 runners from the RG had a TnI level that reached the threshold for diagnosis of myocardial necrosis (50 ng/l). For TnT the proportion was higher, with 10 out of 50 participants reaching this threshold (30 ng/l). No participant had subsequent symptoms suggestive of cardiac disease. This suggests that troponin I levels up to 230 ng/l and troponin T levels up to 210 ng/l (the highest values we recorded) may be physiological in runners completing an endurance event.

    Premarathon BNP levels did not differ between runners with or without SHD. Both groups showed an increase in the mean value of BNP postrace. This was statistically significant in runners with no SHD. Despite this, the absolute units were very small and the clinical significance of the changes unclear. This may indicate that BNP, where clinically indicated, should be interpreted in the normal manner for any runner attending the emergency department. It is difficult to draw conclusions about runners with SHD due to the small sample size.

    When comparing the NSHD and SHD subgroups no difference was shown, either in terms of the median rise in troponin or the proportion of runners with a detectable troponin. However, with only five patients in the SHD group, no firm conclusions can be made and this study should be repeated with a larger sample size for those with SHD.

    The second part of the study looked at those athletes who collapsed and were admitted to the medical tents at the finish. The only other study to look at this13 did not use a control group and only studied those that had collapsed. The authors compared the proportion with a detectable troponin from their research with the proportion of those that completed the race without any medical event from two other studies performed at former Boston marathons.23 ,24 As we had a control group we were able to directly compare and statistically analyse the two groups, avoiding the confounders that arise from comparing figures from marathons of different years. We were able to show that the CG, when compared with the RG, showed a greater proportion of participants with a detectable troponin and a greater degree of rise of troponin. A trend towards higher values of the most sensitive assay (HSTnT) was seen in the CG but did not reach statistical significance. Contrary to Siegel's work13 our study reveals a possible link between cardiac dysfunction and collapse. However, whether the cardiac dysfunction is a cause or effect of collapse is yet to be determined. A brief analysis of the medical records of runners from the CG revealed presenting symptoms including nausea, hypotension and hyperthermia, with no one diagnosis being correlated to the higher levels of troponin.

    Limitations

    The second hypothesis could not be conclusively tested due to the small numbers in the SHD subgroup. We have highlighted the difficulties in recruiting this patient population. A potential solution is to identify these runners through the registration records that hold all participants’ medical information. We also had relatively small numbers in our group of collapsed runners. This again is a challenging group to recruit as when athletes collapse they are usually confused without full mental capacity. In future studies, employment of more data collectors may help.

    Implications

    The data from this research provides information for clinicians treating runners with pre-existing cardiovascular disease or collapsed runners and will allow appropriate interpretation of cardiac biomarker levels. Further research is needed to add to this data and address the limitations of this study.

    Conclusion

    In the introduction we presented three hypotheses about how cardiac biomarkers might respond to strenuous exercise in non-elite runners. The study confirmed our first hypothesis that endurance exercise is associated with an increase in troponin levels in non-elite athletes. Although a proportion of participants had levels above that used to diagnose myocardial necrosis the underlying process appears to be physiological. Our second hypothesis that runners with known heart disease have a greater increase in cardiac biomarkers than those with no known disease was not demonstrated. This may be due to the small sample size of the SHD group. Although a trend towards an increase in HSTnT levels was seen in collapsed versus recruited athletes the results did not reach statistical significance. As such our third hypothesis remains unproven. This could be a consequence of the small number of collapsed athletes studied and a larger sample size may clarify this further.

    This study quantifies the cardiac biomarker levels that can be expected in healthy non-elite runners on completion of a marathon and aids clinical assessment by the emergency team. This is particularly important given the increasing numbers of non-elite participants, including those with cardiac disease, in such endurance events. In addition, further research is needed to study changes in cardiac biomarkers seen in non-elite runners with exercise-associated collapse.

    Acknowledgments

    The authors thank Dr James Thing and Dr William Elson for their help with data collection.

    References

    1. ↵
      1. Fortescue EB,
      2. Shin AY,
      3. Greenes DS,
      4. et al
      . Cardiac troponin increases among runners in the Boston Marathon. Ann Emerg Med 2007;49:137–43.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. George K,
      2. Whyte G,
      3. Stephenson C,
      4. et al
      . Postexercise left ventricular function and cTnT in recreational marathon runners. Med Sci Sports Exerc 2004;36:1709–15.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Herrmann M,
      2. Scharhag J,
      3. Miclea M,
      4. et al
      . Post-race kinetics of cardiac troponin T and I and N-terminal pro-brain natriuretic peptide in marathon runners. Clin Chem 2003;49:831–4.
      OpenUrlFREE Full Text
    4. ↵
      1. Middleton N,
      2. Shave R,
      3. George K,
      4. et al
      . Impact of repeated prolonged exercise bouts on cardiac function and biomarkers. Med Sci Sports Exerc 2007;39:83–90.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Neilan TG,
      2. Januzzi JL,
      3. Lee-Lewandrowski E,
      4. et al
      . Myocardial Injury and ventricular dysfunction related to training levels among non-elite participants in the Boston marathon. Circulation 2006;114:2325–33.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Ohba H,
      2. Takada H,
      3. Musha H,
      4. et al
      . Effects of prolonged strenuous exercise on plasma levels of atrial natruiretic peptide and brain natruiretic peptide in healthy men. Am Heart J 2001;141:751–8.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Scharhag J,
      2. Urhausen A,
      3. Schneider G,
      4. et al
      . Reproducibility and clinical significance of exercise-induced increases in cardiac troponins and N-terminal pro-brain natriuretic peptide (NT-proBNP) in endurance athletes. Eur J Cardiovasc Prev Rehabil 2006;13:388–97.
      OpenUrlPubMedWeb of Science
    8. ↵
      1. Siegal AJ,
      2. Lewandrowski EL,
      3. Chun KY,
      4. et al
      . Changes in cardiac markers including B-natriuretic peptide in runners after the Boston marathon. Am J Cardiol 2001;88:920–3.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Tulloch L,
      2. Robinson D,
      3. Patel A,
      4. et al
      . Raised troponin T and echocardiographic abnormalities after prolonged strenuous exercise—the Australian Ironman Triathlon. Br J Sports Med 2006;40:605–9.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Vidotto C,
      2. Tschan H,
      3. Atamaniuk J,
      4. et al
      . Responses of N-terminal pro-brain natriuretic peptide (NT-proBNP) and cardiac troponin I (cTnI) to competitive endurance exercise in recreational athletes. Int J Sports Med 2005;26:645–50.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Whyte G,
      2. George K,
      3. Sharma S,
      4. et al
      . Cardiac fatigue following prolonged endurance exercise of differing distances. Med Sci Sports Exerc 2000;32:1067–72.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Jassal DS,
      2. Moffat D,
      3. Krahn J,
      4. et al
      . Cardiac injury markers in non-elite runners. Int J Sports Med 2009;30:75–9.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Siegel AJ,
      2. Januzzi J,
      3. Sluss P,
      4. et al
      . Cardiac biomarkers, electrolytes, and other analytes in collapsed marathon runners: implications for the evaluation of runners following competition. Am J Clin Pathol 2008;129:948–51.
      OpenUrlAbstract/FREE Full Text
    14. ↵
    15. ↵
      1. Aagaard P,
      2. Sahlen A,
      3. Braunschweig F
      . Performance trends and associations with biomarkers in a 30 km cross-country race. Med Sci Sports Exerc 2011;44:894–9.
      OpenUrl
    16. ↵
      1. Roberts WO
      . Exercise-associated collapse care matrix in the marathon. Sports Med 2007;37:431–3.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Roberts WO
      . A 12-yr profile of medical injury and illness for the Twin Cities Marathon. Med Sci Sports Exerc 2000;32:1549–55.
      OpenUrlPubMedWeb of Science
    18. ↵
      1. Holtzhausen LM,
      2. Noakes TD
      . The prevalence and significance of post-exercise (postural) hypotension in ultramarathon runners. Med Sci Sports Exerc 1995;27:1595–601.
      OpenUrlPubMedWeb of Science
    19. ↵
      1. Eijsvogels T,
      2. George K,
      3. Shave R,
      4. et al
      . Effect of prolonged walking on cardiac troponin levels. Am J Cardiol 2010;105:267–72.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Maisel A
      . The coming of age of natriuretic peptides: the emperor does have clothes! J Am Coll Cardiol 2006;47:61–4.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Reed MJ,
      2. Newby DE,
      3. Coull AJ,
      4. et al
      . Role of brain natriuretic peptide (BNP) in the risk stratification of adult syncope. Emerg Med J 2007;24:769–73.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Scharhag J,
      2. George K,
      3. Shave R,
      4. et al
      . Exercise-associated increases in cardiac biomarkers. Med Sci Sports Exerc 2008;40:1408–15.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Neilan T,
      2. Januzzi J,
      3. Lee-Lewandrowski E,
      4. et al
      . Myocardial injury and ventricular dysfunction related to training levels among nonelite participants in the Boston Marathon. Circulation 2006;114:2325–33.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Fortescue E,
      2. Shin A,
      3. Greenes D,
      4. et al
      . Cardiac troponin increases among runners in the Boston Marathon. Ann Emerg Med 2007;49:137–43.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors PB was responsible for the design, day-to-day running of the study, data analysis and manuscript writing. SLD contributed greatly to all stages already mentioned and was involved in the main changes to the article. JL and DM helped in data collection and in proof reading and writing of the manuscript. SB and AR helped in the statistical analysis, proofreading and writing of the manuscript. TH was instrumental in the study design and provided much needed support and feedback at all stages of the study. The authors also contributed heavily to writing and reviewing the manuscript.

    • Funding The statistical analysis was funded by the Royal London Hospital.

    • Competing interests None.

    • Ethics approval East London and City Alpha research ethics committee.

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

    • Data sharing statement Additional unpublished data is only available to PB.