Share this article on:

Repolarization Perturbation and Hypomagnesemia after Extreme Exercise


Medicine & Science in Sports & Exercise: September 2012 - Volume 44 - Issue 9 - p 1637–1643
doi: 10.1249/MSS.0b013e318258aaf4
Clinical Sciences

Purpose Strenuous exercise induces significant increases in inflammatory and cardiac biomarkers and transient dysfunction of the left ventricle. It is still unclear whether the electrophysiological correlate of these alterations can also be observed in ECG recordings, which indicate increased vulnerability for arrhythmias.

Methods ECG parameters were measured and compared with inflammatory and electrolyte statuses in 198 healthy men (42 ± 9 yr) 1 wk before (baseline) and at 0, 24, and 72 h after participating in a marathon.

Results HR-corrected QT interval (QTc) duration increased significantly immediately after the race (442.4 ± 23.0 ms) compared with baseline (415.3 ± 22.5 ms, P < 0.001) and returned to baseline values within 72 h (415.8 ± 24.7 ms). Other indices of ventricular repolarization (Sagie–Framingham-corrected QTcF, T peakT end (T pe) duration, T pe/QT ratio) showed similar results.

Compared with baseline, significant decreases in serum concentrations of magnesium and potassium were shown immediately after the race (median (interquartile range): [Mg2+] = 0.85 (0.79–0.92) vs 0.77 (0.70–0.82) mmol·L−1, P < 0.001; [K+] = 4.27 (4.02–4.58) vs 4.14 (3.81–4.57) mmol·L−1, P < 0.05). Interleukin 6 values were significantly augmented immediately after the race compared with baseline (interleukin 6 = 2.08 (1.96–2.20) vs 30.56 (19.83–43.63) ng·L−1, P < 0.001) and returned to baseline values within 72 h.

Significant associations were not observed between ECG alterations and inflammatory or electrolyte concentrations (all P > 0.05).

Conclusions Cardiac repolarization was significantly altered immediately after a marathon, coincident with hypomagnesemia and hypokalemia. Inflammatory and electrolyte statuses returned to baseline values within 72 h. The current data do not support that ECG alterations after marathon running represent an increased risk for arrhythmic events. However, further investigation is warranted to describe these relationships in more detail.

Supplemental digital content is available in the text.

1Department of Prevention and Sports Medicine, Klinikum rechts der Isar, Technische Universitaet Muenchen, Munich, GERMANY; 2Department for Medical Statistics and Epidemiology, Klinikum rechts der Isar, Technische Universitaet Muenchen, Munich, GERMANY; and 3Munich Heart Alliance, Munich, GERMANY

Address for correspondence: Johannes Scherr, M.D., Department of Prevention and Sports Medicine, Klinikum rechts der Isar, Technische Universitaet Muenchen, Connollystr. 32, D-80809 Munich, Germany; E-mail:

Submitted for publication January 2012.

Accepted for publication March 2012.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (

Regular moderate physical activity (PA) is an important preventive strategy to improve cardiovascular risk factors, delay cardiovascular disease, and decrease cardiovascular mortality (17,39). However, although contradictory (30), there is evidence suggesting that the performance of strenuous exercise such as marathon running may increase the risk of sudden cardiac death (SCD) (1,23,39). The primary causes for exercise-related SCD are malignant arrhythmias, which may be caused by numerous cardiac anomalies (24).

Extreme exercise increases the concentrations of biomarkers of cardiac strain and cardiac injury such as cardiac troponins (cTn) as well as inflammatory markers (e.g., interleukin 6 (IL-6)) (32). Systemic IL-6 levels have been shown to be an independent predictor of sudden death in asymptomatic middle-age European men (13). These proarrhythmogenic effects of IL-6 and other inflammatory markers have also been described in patients with atrial fibrillation, ventricular tachycardia, and fibrillation (4,19,36). Electrolyte disturbances (particularly hypokalemia and hypomagnesemia) are also known to induce cardiac arrhythmias and increase the incidence of SCD. Hollifield (18) showed that diuresis-induced hypokalemia and hypomagnesemia are linked to increased ventricular ectopy. Furthermore, in a review by Chiuve et al. (7), both plasma concentrations and dietary intake of magnesium were strongly and inversely related to risk of SCD in adult women.

Besides the effects of clinical chemistry on markers described above, strenuous exercise has also been implicated in transient functional as well as persistent structural myocardial alterations (e.g., left and right ventricular dysfunction) and even myocardial fibrosis (3,26). Benito et al. (2) have shown cardiac remodeling with myocardial fibrosis in the right ventricle and both atria and increased arrhythmia in rats after a long-term strenuous exercise. However, of the few articles that have investigated electrophysiological changes after participation in a prolonged and strenuous exercise, the results are inconsistent (25,31,34).

Therefore, we investigated ECG changes before and after a strenuous exercise and their relationship with changes in electrolyte and inflammatory marker concentrations in a large cohort of marathon runners up to 72 h after the race.

Back to Top | Article Outline


Study Design

This prospective study was conducted with the ethical approval of the Ethics Committee of the Technical University of Munich Medical School. All participants gave written informed consent, and the study was conducted according to the Declaration of Helsinki.

Back to Top | Article Outline


Male patients who were (at screening) between 20 and 60 yr, had previously successfully completed at least one-half marathon, intended to participate at the 2009 Munich Marathon, and submitted a written informed consent were included.

Patients with the following characteristics were excluded: with known cardiac disease, taking pharmaceutical treatment for diabetes mellitus or arterial hypertension, with musculoskeletal or psychiatric disease, with neoplasia, with acute or chronic infection or inflammatory disease, with known malabsorption, taking medications or supplements influencing immune function, and with a history of alcohol and/or drug abuse or addiction.

Back to Top | Article Outline


Participants were screened to assess inclusion and exclusion criteria 4–5 wk before the marathon (V 1).

Baseline data were collected during the week before the race (V 2), including training history questionnaires, physical examination, anthropometry, clinical chemistry, and electrocardiogram.

Collection of blood samples, assessment of blood pressure, and ECG were all performed within 30 min after finishing the race (V 3).

Follow-up examinations (ECG, blood samples, and blood pressure) were performed at 24 h (V 4) and 72 h (V 5) after the marathon race in identical settings.

During the marathon, all participants were asked to use an HR monitor to determine the individual exercise intensity. %HRmax was calculated as a ratio of mean HR during marathon (HRM) and maximum HR (HRmax) as calculated by the following formula: HRmax = 208 − 0.7 × age (yr) (38). One hundred fifty-one participants (76%) wore an HR monitor during the race.

Participants documented all medication taken in the 2 wk before the marathon (e.g., nonsteroidal anti-inflammatory drugs).

Body mass index was calculated as the ratio of weight and the square of height (kg·m−2). Total body fat was assessed by the seven-site skinfold caliper technique using the formula of Jackson and Pollock (20). Hypertension was defined as having >140 mm Hg systolic and/or >90 mm Hg diastolic blood pressure (8). An elevated cholesterol level was defined as total cholesterol >240 mg·dL−1. Smoking was defined as current smoking or having regularly smoked within the previous year.

Back to Top | Article Outline

ECG analyses.

Electrocardiography was performed using standard 12-lead placement and equipment (custo cardio 200 with custo diagnostics 3.8.3; custo med GmbH, Ottobrunn, Germany). ECG recordings of duration and amplitude were critically studied visually by one of three study physicians (J.S., A.P., and M.H.). Standard 12-lead ECGs were performed after 5 min of rest in a supine position and were digitally recorded for a duration of 10 s, with a speed of 50 mm·s−1 and a voltage scale equivalent of 10 mm·mV−1. All investigators conducting the ECG recordings were trained to ensure consistent and exact positioning of the electrodes. All ECG patterns were evaluated according to standard clinical criteria (35). No prerace ECG abnormalities that warranted immediate medical attention or contraindicated participation in the marathon race were found. All ECG reports were evaluated for quality and irregularities by two of the three physicians, including at least one cardiologist. All ECG investigators were blinded, and ECG analyses from all three investigators were tested for interobserver consistency. In cases in which there was disagreement in the analyses (e.g., regarding uncommon ECG findings), the final ECG interpretation was determined by a majority consensus (a two-thirds majority).

Normal ECG changes due to physical exertion were distinguished from suspicious abnormalities according to current guidelines (10,11). The absolute values of the following ECG parameters were investigated and described: HR, P-wave duration (onset was defined as a positive deflection after the T wave preceding the QRS waveform and deviating from the isoelectric line composed of the TP interval; offset was determined as the return to isoelectric baseline immediately before the QRS waveform), PQ interval duration, QRS duration [defined as the interval from the onset of the Q or R wave (junction between the isoelectric line and the beginning of the Q or R deflection) to the end of the R or S wave (junction with the ST segment)], QT and QTc interval [QT interval was analyzed in as many of the 12 leads as possible, defined as the longest period between the onset of the QRS complex and the end of the T wave, whereas the points were analyzed independently of the leads; QTc interval calculated by Bazett’s method (QTc = QT /√RR interval) in leads II, V 2, and V 6], which is by far the most commonly used method to derive HR-corrected QT interval (QTc), and by Sagie–Framingham’s method (QTcF = QT + 0.154 (1 − RR)), which is less HR dependent at HR >60 bpm than Bazett’s corrected QTc (6), JTc interval (calculated by subtracting the QRS duration from the QTc interval in leads II, V 2, and V 6), duration from the peak to the end of the T wave (T pe) in the precordial leads (defined as the interval from the peak of a positive T-wave or the nadir of a negative T-wave to the end of the T-wave), and the T pe/QT ratio (calculated using the corresponding values from each lead). T pe is considered to reflect the transmural axis of the left ventricle, therefore providing an index of transmural dispersion of repolarization (21,22). A prolonged T pe interval and an increased T pe/QT ratio are considered to be noninvasive markers of arrhythmogenesis because they have been linked with the development of malignant ventricular arrhythmia (16,21,22).

Back to Top | Article Outline

Blood samples.

Fasting blood samples were drawn from an antecubital vein with subjects in a supine position at visits 1, 2, 4, and 5. The blood collection directly after the race (V 3) was not performed in a fasted state. All participants were instructed to refrain from long runs and strenuous exercise for at least 3 d before the prerace blood draws (V 2). Because albumin concentration increased significantly from before to after the race, all dehydration-dependent concentrations were corrected for changes in plasma volume as previously described (12).

Back to Top | Article Outline


This was measured using a solid-phase, two-site chemiluminescent immunometric assay with the Immulite system (Siemens Healthcare, Eschborn, Germany). Expected values in healthy individuals range from nondetectable to 5.9 ng·L−1 for IL-6. The analytical sensitivity is 2 ng·L−1 for IL-6. The measuring range is up to 1000 ng·L−1 for all measured parameters.

Back to Top | Article Outline

Plasma electrolytes.

Potassium and sodium were measured with an ion-selective electrode (AU 2700; Olympus/Beckman Coulter, Nyon, Switzerland). Magnesium and calcium were measured photometrically (AU 2700; Olympus/Beckman Coulter). Expected values in healthy individuals range from 135 to 150 mmol·L−1 for sodium (hyponatremia [Na+] < 135 mmol·L−1), from 3.5 to 5.0 mmol·L−1 for potassium (hypokalemia [K+] < 3.5 mmol·L−1), from 2.0 to 2.65 mmol·L−1 for calcium, and from 0.78 to 1.03 mmol·L−1 for magnesium (hypomagnesemia [Mg2+] < 0.78 mmol·L−1). The interassay coefficient of variation under actual routine conditions is 0.76% at a potassium concentration of 4.8 mmol·L−1, 0.64% for sodium at a concentration of 138 mmol·L−1, 1.09% for magnesium at a concentration of 1.09 mmol·L−1, and 0.46% for calcium at a concentration of 2.79 mmol·L−1.

Back to Top | Article Outline

Statistical analysis.

Data analysis was performed using PASW Statistics 18.0.2 (SPSS, Inc., Chicago, IL). Quantitative statistics are described as means with their SD and ranges (for normally distributed data) or medians with interquartile ranges (IQR; for non–normally distributed data; IQR = 25th–75th percentile). Assumption of normal distribution of data was verified by using descriptive methods (skewness, outliers, and distribution plots) and inferential statistics (Shapiro–Wilk test).

Because of the non–normally skewed distribution of some outcome parameters (some clinical chemistry data), natural logarithm transformation was applied before parametric data analysis (linear regression). Back transformation of regression coefficients (using a simple exponential function) gives an estimate for the median relative change of the outcome measure per a one-unit increment of the corresponding explanatory variable. Simple correlation analysis was performed using Spearman rank correlation (ρ).

Wilcoxon signed rank test was used to evaluate changes in serum biomarkers and ECG markers between two time points. P < 0.05 was considered to indicate statistical significance, and Bonferroni correction of P values was applied variable-wise within any multiple comparison. Testing was performed two-sided.

Back to Top | Article Outline


Participants’ characteristics.

During the 4-month recruitment phase, 277 participants were enrolled and followed up for 2 months after marathon participation. This follow-up period was used to monitor for potential adverse clinical cardiovascular events (e.g., cardiac syncope). ECG reports for all four visits were able to be analyzed in 198 participants (71%). Reasons for dropout or exclusion were intake of nonsteroidal anti-inflammatory drugs (n = 37), failure to finish the marathon race (n = 15), common cold (n = 7), orthopedic problems (n = 5), gastrointestinal problems (n = 3), personal time constraints (n = 9), and others (n = 3). Subjects’ baseline characteristics are presented in Table 1.



Back to Top | Article Outline

Alterations of ECG values before versus after marathon.

Several suspicious ECG abnormalities were observed; most of them were detected during the repolarization phase as represented in Table 2. Most of these abnormalities occurred in QTc and JTc duration at V 3 (after the race) and V 4 (24 h after the race, all P < 0.001; Fig. 1). These changes were no longer existent at V 5 (72 h after the race). These changes were also seen using the Sagie–Framingham correction method (QTcF; Table 2).





We observed a significant increase (all P < 0.001) in prolonged HR-corrected QT intervals between V 2 (12.5% of the participants, 1 wk before the race), V 3 (48.2%, immediately after the race), and V 4 (48.2%, 24 h after the race). These changes returned to baseline within 72 h (V 5).

There were four participants with QTc prolongation (QTc intervals ≥ 500 ms and therefore suspicious for long QT syndrome), which persisted for 24 h after the race (V 4). These participants did not differ in resting HR from the other participants (60 ± 2 vs 58 ± 10 bpm, P = 0.46). However, a tendency toward clinically relevant longer QTc durations was observed at V 2 (434 ± 31 vs 415 ± 23 ms, P = 0.23). Regarding the parameters of clinical chemistry, these participants had higher IL-6 at V 3 than the other participants (71.5 ± 29.1 vs 35.6 ± 26.0 ng·L−1, P = 0.02). Neither training history nor fluid intake during the marathon had an influence on QTc interval abnormalities or serum electrolytes (all P > 0.05). Detailed data of these four participants are represented in Table 3. To illustrate, the ECGs from all visits of one of the participants with QTc prolongation at V 4 are shown in the supplementary material (see Figure, Supplemental Digital Content 1,



Further ECG parameters representing the spatial dispersion of ventricular repolarization (e.g., T peak-to-end (T pe) interval and the T pe/QT ratio) also showed significant changes after the race with normalization within 72 h (Table 2).

Back to Top | Article Outline

Uncommon ECG findings.

In additions to the changes observed for ventricular repolarization described above, we observed more participants with signs of atrial enlargement immediately after the race compared with that before the race (left atrial enlargement = 17.9% at V 2 vs 29.0% at V 3; right atrial enlargement = 3.6% at V 2 vs 17.7% at V 3, both P < 0.001). These enlargements were reversed within 24 h. No other relevant uncommon ECG findings were observed.

Except for one participant, who had an initial diagnosis of a third-degree atrioventricular block immediately after the race, no significant ECG pathologies were observed. The patient with the third-degree atrioventricular block received further diagnostic tests in accordance with the current guidelines. No adverse clinical events were observed during the follow-up period.

Back to Top | Article Outline

Electrolyte and inflammatory status.

We observed significant changes in all investigated serum electrolyte concentrations as shown in Table 4. Of the electrolytes, magnesium concentrations showed the greatest decreases due to marathon running, with 57.5% of the subjects having hypomagnesemia immediately after the race (compared with 23.0% before the race, P < 0.001; Fig. 2). Concomitant to the hypomagnesemia, a decrease in serum potassium level was observed immediately after the race compared with that before the race (P < 0.05). Both of these conditions were reversed within 24 h. Hypernatremia (serum Na < 135 mmol·L−1) was not observed in any participant at before or after the race visits. Plasma volume decreased an average of 3.5% in marathon participants.





There were significant changes in IL-6 between the first three examinations (before the race, immediately after the race, and 24 h after the race; Table 4). At the prerace visit and the visits 24 and 72 h after the race, IL-6 concentrations of most of the subjects were below the lower limit of detection. Measurable IL-6 levels were only detectable at V 3 and were significantly increased compared with baseline (P < 0.001).

Multiple regressions analysis revealed no significant associations between changes in ECG markers of repolarization (QTc, T pe duration, T pe/QT ratio) and changes in serum electrolytes and inflammatory markers between V 2 and V 3 (all P > 0.05). There were also no significant associations using simple correlation analysis (all P > 0.05, Spearman ρ ranging from 0.003 to 0.08). Furthermore, neither increases in inflammatory markers nor changes in serum electrolytes were associated with changes in repolarization. Similarly, finishing time and exercise intensity were not associated with ECG changes (P > 0.05).

Back to Top | Article Outline


This study is the first to simultaneously analyze ECG and laboratory parameters in relation to marathon running in a very large cohort. We have shown that marathon running induces changes both in serum electrolytes and in inflammatory markers and, simultaneously, significantly alters cardiac repolarization. The latter may represent cardiac strain and might be related to an increased risk of arrhythmic events.

Currently, there are three other studies dealing with marathon-induced ECG changes. In the most recent study, Minns et al. (25) investigated ECG changes in 87 half-marathon runners from before to immediately after the race. They found abnormalities in 62% of the runners. The most common abnormalities were atrial enlargement, conduction abnormalities, new Q waves, and nonspecific ST/T wave changes. However, they did not describe any changes in repolarization.

The first study examining marathon-induced ECG changes was published in 1989 (34). In this study, 31 marathon runners were investigated on the day before the race, directly after the race, and 7–14 d after the race with signal-averaged ECG (SAECG). They observed no abnormal SAECG findings after the race and even reported improvements on the initial abnormal SAECG.

In a third study, Sahlen et al. (31) investigated ventricular repolarization with vectorcardiography before the race and immediately after and 24 h and 6 d after the race in 15 older male long-distance runners. They observed—in accordance to our results—an increase in QTc and T pe intervals. Importantly, in contrast to our investigation, they only examined a limited number of runners (n = 15). Furthermore, the methods used by Sahlen et al. clearly increased the complexity and cost of the examination. Therefore, our current study is the first to comprehensively describe the results of a widely applicable and easy-to-use examination method in a very large cohort.

There are several studies reporting an increased risk of SCD during strenuous exercise (1,14); however, the theories attempting to explain this phenomenon are speculative (30). Until now, no plausible explanation of this phenomenon has been advanced. A likely mechanism—especially in younger athletes without structural cardiac pathologies—may have its origin in previously undetected malignant arrhythmias (29). As observed in the present study, there is an increased incidence of ECG markers of arrhythmogenic events with altered ventricular repolarization that coincides with marathon running. These markers are (a) prolongation of QTc duration (33) and (b) T pe duration and T pe/QT ratio (16,22). These results are similar to those seen in patients with stable coronary artery disease showing that the spatial dispersion of repolarization is increased during exercise therefore increasing the risk of arrhythmogenic events (21).

Among others, the percentage of subjects with a QT duration suggestive for a long QT syndrome (LQTS: QT duration > 500 ms) was significantly increased. Furthermore, a significant prolongation of the T peakT end interval was observed after the race. This has been shown to be associated to an increased disposition to develop malignant ventricular tachycardia (e.g., Torsades de pointes tachycardia in subjects with LQTS) and SCD (22,28,37).

These electrophysiological alterations are aggravated by the fact that several changes in serum electrolytes were concomitantly observed. Decreases in magnesium and potassium concentrations can cause hyperexcitability of myocytes and facilitate arrhythmias (15). However, causality between altered ventricular repolarization and changes in serum electrolyte concentrations has not been clearly shown in the present study.

As described by Chiang (5), it is possible that prolongation of ventricular repolarization could be due to a dysregulation of the autonomic nervous system with enhanced sympathetic activity occurring during and immediately after the race. Because of these observations, it may be concluded that the prolongation of the QTc interval is primarily caused by increased sympathetic activity rather than electrolyte disturbances. This “hypersympathetic” environment with release of catecholamines is likely a central component to exercise-related cardiac dysrhythmia and increases the risk of malignant arrhythmias and SCD (9).

A direct association between increased inflammatory markers and alterations in resting ECG immediately or 24 h after the race was not observed. However, participants with QTc durations suggestive of LQTS after the race had significantly higher IL-6 levels after the race compared with all other participants. Therefore, the suggested association between proarrhythmogenic effects of IL-6 and ventricular arrhythmic events cannot be excluded (4,13,36).

Back to Top | Article Outline


First, we examined ECG after the race only for a duration of 2 min, and no continuous ECG monitoring (e.g., Holter monitoring) was performed during marathon running. Furthermore, we were not able to observe any potential malignant arrhythmias. Therefore, we cannot draw conclusions regarding arrhythmias during marathon running, and we are limited to describing alterations observed between baseline and after the race. Although there are several difficulties involved in performing such investigations, future studies should attempt to record and analyze ECG continuously during long-term strenuous exercise such as marathon running. However, the investigations conducted in this study (resting ECG) are comparable to those that are typically conducted in collapsed or injured athletes admitted to hospitals during and after marathon races, making the present results applicable to common clinical settings.

In addition, we did not perform direct measurements of sympathetic activity (e.g., plasma noradrenaline concentrations) and therefore cannot show a direct association between activity of the autonomic nervous system and ECG changes. However, it is well known that mean plasma levels of catecholamines increase significantly during marathon participation (27). Therefore, it seems likely that both sympathetic withdrawal and increased vagal tone after the race, combined with large quantities of catecholamines still circulating in the cardiovascular system as a result of intensive sympathetic stimulation, may contribute to the variation of ventricular repolarization.

Back to Top | Article Outline


In marathon runners, there is altered ventricular repolarization with increased indexes of arrhythmic risk accompanied by hypomagnesemia, acute inflammatory response, and a decrease in serum potassium levels after the race. These changes might contribute to an increased susceptibility for arrhythmogenic events during strenuous and prolonged exercise.

Erdinger Weissbraeu, Werner Brombach GmbH, provided financial support for this study but had no direct role in the study’s design, conduct, analysis, and interpretation of data and reporting beyond approval of the scientific protocol in peer-review for funding. No other grants were received. ID: NCT00933218. The authors would like to thank the staff and especially the doctoral students (Jana Habermann, Charlotte Hartmann, and Esther Seifert) of the Department of Prevention and Sports Medicine, Technische Universitaet Muenchen for their assistance with this project.

None of the authors had any personal or financial conflicts of interest.

The study followed the guidelines on good publication practice (GPP2).

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Albert CM, Mittleman MA, Chae CU, Lee IM, Hennekens CH, Manson JE. Triggering of sudden death from cardiac causes by vigorous exertion. N Engl J Med. 2000; 343 (19): 1355–61.
2. Benito B, Gay-Jordi G, Serrano-Mollar A, et al.. Cardiac arrhythmogenic remodeling in a rat model of long-term intensive exercise training. Circulation. 2011; 123 (1): 13–22.
3. Breuckmann F, Mohlenkamp S, Nassenstein K, et al.. Myocardial late gadolinium enhancement: prevalence, pattern, and prognostic relevance in marathon runners. Radiology. 2009; 251 (1): 50–7.
4. Chen Y, Zhang Q, Liao YH, et al.. Effect of tumor necrosis factor-alpha on neutralization of ventricular fibrillation in rats with acute myocardial infarction. Mediators Inflamm. 2011; 2011: 565238.
5. Chiang CE. Congenital and acquired long QT syndrome. Current concepts and management. Cardiol Rev. 2004; 12 (4): 222–34.
6. Chiladakis J, Kalogeropoulos A, Arvanitis P, Koutsogiannis N, Zagli F, Alexopoulos D. Heart rate—dependence of QTc intervals assessed by different correction methods in patients with normal or prolonged repolarization. Pacing Clin Electrophysiol. 2010; 33 (5): 553–60.
7. Chiuve SE, Korngold EC, Januzzi JL Jr, Gantzer ML, Albert CM. Plasma and dietary magnesium and risk of sudden cardiac death in women. Am J Clin Nutr. 2011; 93 (2): 253–60.
8. Chobanian AV, Bakris GL, Black HR, et al.. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003; 42 (6): 1206–52.
9. Corrado D, Migliore F, Basso C, Thiene G. Exercise and the risk of sudden cardiac death. Herz. 2006; 31 (6): 553–8.
10. Corrado D, Pelliccia A, Bjornstad HH, et al.. Cardiovascular pre-participation screening of young competitive athletes for prevention of sudden death. Eur Heart J. 2005; 26 (5): 516–24.
11. Corrado D, Pelliccia A, Heidbuchel H, et al.. Recommendations for interpretation of 12-lead electrocardiogram in the athlete. Eur Heart J. 2010; 31 (2): 243–59.
12. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol. 1974; 37 (2): 247–8.
13. Empana JP, Jouven X, Canoui-Poitrine F, et al.. C-reactive protein, interleukin 6, fibrinogen and risk of sudden death in European middle-aged men: the PRIME study. Arterioscler Thromb Vasc Biol. 2010; 30 (10): 2047–52.
14. Ferreira M, Santos-Silva PR, de Abreu LC, et al.. Sudden cardiac death athletes: a systematic review. Sports Med Arthrosc Rehabil Ther Technol. 2010; 2: 19.
15. Fiset C, Kargacin ME, Kondo CS, Lester WM, Duff HJ. Hypomagnesemia: characterization of a model of sudden cardiac death. J Am Coll Cardiol. 1996; 27 (7): 1771–6.
16. Gupta P, Patel C, Patel H, et al.. T(p-e)/QT ratio as an index of arrhythmogenesis. J Electrocardiol. 2008; 41 (6): 567–74.
17. Haskell WL, Lee IM, Pate RR, et al.. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007; 116 (9): 1081–93.
18. Hollifield JW. Electrolyte disarray and cardiovascular disease. Am J Cardiol. 1989; 63 (4): 21B–6.
19. Issac TT, Dokainish H, Lakkis NM. Role of inflammation in initiation and perpetuation of atrial fibrillation: a systematic review of the published data. J Am Coll Cardiol. 2007; 50 (21): 2021–8.
20. Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr. 1978; 40 (3): 497–504.
21. Korantzopoulos P, Letsas KP, Christogiannis Z, et al.. Exercise-induced repolarization changes in patients with stable coronary artery disease. Am J Cardiol. 2011; 107 (1): 37–40.
22. Kors JA, Ritsema van Eck HJ, van HG. The meaning of the TpTe interval and its diagnostic value. J Electrocardiol. 2008; 41 (6): 575–80.
23. Marijon E, Tafflet M, Celermajer DS, et al.. Sports-related sudden death in the general population. Circulation. 2011; 124 (6): 672–81.
24. Maron BJ, Doerer JJ, Haas TS, Tierney DM, Mueller FO. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980–2006. Circulation. 2009; 119 (8): 1085–92.
25. Minns AB, McFarland C, Strachan M, et al.. Electrocardiogram and echocardiogram findings in runners completing a half marathon. Am J Emerg Med. 2011; 29 (9): 1182–7.
26. Neilan TG, Yoerger DM, Douglas PS, et al.. Persistent and reversible cardiac dysfunction among amateur marathon runners. Eur Heart J. 2006; 27 (9): 1079–84.
27. Ohba H, Takada H, Musha H, et al.. Effects of prolonged strenuous exercise on plasma levels of atrial natriuretic peptide and brain natriuretic peptide in healthy men. Am Heart J. 2001; 141 (5): 751–8.
28. Panikkath R, Reinier K, Uy-Evanado A, et al.. Prolonged T peak to T end interval on the resting electrocardiogram is associated with increased risk of sudden cardiac death. Circ Arrhythm Electrophysiol. 2011; 4 (4): 441–7.
29. Papadakis M, Whyte G, Sharma S. Preparticipation screening for cardiovascular abnormalities in young competitive athletes. BMJ. 2008; 337: a1596.
30. Redelmeier DA, Greenwald JA. Competing risks of mortality with marathons: retrospective analysis. BMJ. 2007; 335 (7633): 1275–7.
31. Sahlen A, Rubulis A, Winter R, et al.. Cardiac fatigue in long-distance runners is associated with ventricular repolarization abnormalities. Heart Rhythm. 2009; 6 (4): 512–9.
32. Scherr J, Braun S, Schuster T, et al.. 72-h kinetics of high-sensitive troponin T and inflammatory markers after marathon. Med Sci Sports Exerc. 2011; 43 (10): 1819–27.
33. Schouten EG, Dekker JM, Meppelink P, Kok FJ, Vandenbroucke JP, Pool J. QT interval prolongation predicts cardiovascular mortality in an apparently healthy population. Circulation. 1991; 84 (4): 1516–23.
34. Smith GS, Vacek JL, Wilson DB, Hawkins JW, Boyer TA. Exercise-induced alterations of signal-averaged electrocardiograms in marathon runners. Am Heart J. 1989; 118 (6): 1198–202.
35. Sokolow M, Freidlander RD. The normal unipolar precordial and limb lead electrocardiogram. Am Heart J. 1949; 38 (5): 665–87.
36. Streitner F, Kuschyk J, Veltmann C, et al.. Role of proinflammatory markers and NT-proBNP in patients with an implantable cardioverter–defibrillator and an electrical storm. Cytokine. 2009; 47 (3): 166–72.
37. Takenaka K, Ai T, Shimizu W, et al.. Exercise stress test amplifies genotype–phenotype correlation in the LQT1 and LQT2 forms of the long-QT syndrome. Circulation. 2003; 107 (6): 838–44.
38. Tanaka H, Monahan KD, Seals DR. Age-predicted maximal heart rate revisited. J Am Coll Cardiol. 2001; 37 (1): 153–6.
39. Thompson PD, Franklin BA, Balady GJ, et al.. Exercise and acute cardiovascular events placing the risks into perspective: a scientific statement from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism and the Council on Clinical Cardiology. Circulation. 2007; 115 (17): 2358–68.


©2012The American College of Sports Medicine