American Journal of Emergency Medicine (2011) 29, 1182-1187

Brief Report

Electrocardiogram and echocardiogram findings in runners completing a half marathon

Alicia B. Minns MD^a,b,?, Cameron McFarland MD^b, Monet Strachan RDCS^c, Wendy Austin MD^c, Edward Castillo MPH, PhD^b, Ori Ben-Yehuda MD^c,

Richard F. Clark MD^a,b

^aDivision of Medical Toxicology, Department of Emergency Medicine, University of California, San Diego, San Diego,

CA 92103-8925, USA

^bDepartment of Emergency Medicine, University of California, San Diego, San Diego, CA 92103-8676, USA

^cDivision of Cardiology, Department of Medicine, University of California, San Diego, San Diego, CA 92103-8411, USA

Received 22 March 2010; revised 11 June 2010; accepted 24 June 2010

Abstract

Objective: The purpose of this study was to prospectively evaluate Electrocardiograms before and after running a half marathon to characterize the changes that occur after exertion. Echocardiograms were also done postrace on selected runners.

Methods: Volunteer runners older than 18 years completed a questionnaire detailing demographic and medical history. Prerace ECGs were performed at a prerace symposium and postrace ECGs were performed within 15 minutes of the participants’ completion of the race. Echocardiograms were performed on a random sampling of runners who were judged to have abnormal or changed postrace ECGs.

Results: Eighty-seven runners were enrolled and completed the study. There were 46 males (53%) and 41 females (47%). Fifty-four (62%) runners had changes noted in their ECGs. The most common changes noted were atrial enlargement (37 runners). Other abnormalities seen on the ECGs included new Conduction abnormalities, new Q waves, nonspecific ST/T-wave changes, and resolution of previous abnormalities that were seen on initial ECG. There was no statistically significant difference between runners with ECG changes and runners without ECG changes when comparing sex, age, previous marathon experience, exercise, history of exercise-induced chest pain, medical history, and family history of heart disease. Twenty runners with changed or abnormal postrace ECGs had postrace echocardiograms, and 8 were abnormal. Of the abnormal echocardiograms, 2 runners had normal but changed postrace ECGs and 4 runners had abnormalities that were inconsistent with what was seen on ECG.

Conclusion: Our study suggests that ECG abnormalities and changes can occur in distance runners, but their significance is unclear.

(C) 2011

* Corresponding author. Tel.: +1 619 733 7312; fax: +1 858 715 6361.

E-mail address: [email protected] (A.B. Minns).

0735-6757/$ - see front matter (C) 2011 doi:10.1016/j.ajem.2010.06.035

Introduction

Marathon running has surged in popularity in the last quarter century, with more than 400 000 runners participa- ting annually in the United States and millions worldwide [1]. Distance runners can have a variety of injuries, most relating to overuse and stress to the lower extremities. However, Cardiac effects are also seen in endurance sports, and sudden cardiac death and related cardiac syndromes have been reported [2].

The 12-lead electrocardiogram can show a spectrum of abnormalities in trained athletes. These abnormal findings are generally considered clinically insignificant and a consequence of athletic conditioning [3]. Of the abnormal ECG patterns described in trained athletes, only a minority will actually display clinically significant structural cardiac disease as diagnosed with echocardiography [4]. Pelliccia et al [5] evaluated 1050 athletes and identified Abnormal ECGs in 40%, with more distinctly abnormal patterns in those engaged in endurance sports such as cycling and cross- country skiing. To our knowledge, ECG changes specifically after distance running has not been studied. Evaluating ECGs in these athletes may further the understanding of the interrelation of intense physical exertion and cardiac structure and function. The purpose of this study was to prospectively evaluate ECGs before and after running a half marathon to characterize the changes that occur after exertion. We also evaluated echocardiograms in selected runners with changes or abnormalities in postrace ECGs.

Methods

This is a prospective observational study of a convenience sample of runners designed to examine the ECG changes that occur after participating in a half marathon (13.1 miles). Echocardiograms were also done on selected runners with postrace ECG changes and correlated with ECG findings. A popular half marathon run in the downtown San Diego area occurring in August of 2008 was chosen. Runners older than 18 years were offered participation in the study at a prerace event 1 to 2 days before the race. After signing a written consent, each runner was given a study number and completed a questionnaire detailing medical history, medica- tions, previous exercise experience, previous marathon experience (both half and full marathon), family history, and if there was a history of exercise-induced chest pain. Exercise experience was arbitrarily categorized as: “never,” “rarely” (once per few weeks), “sometimes” (1-2 times/wk), “often” (3-6 times/wk), and “everyday.” Enrollment was limited to the first 100 runners to consent.

Standard 12-lead ECGs were then performed at the prerace symposium, with the subject in the supine position during quiet respiration, and were recorded at 25 mm/s. All ECG patterns were evaluated according to commonly adopted clinical criteria [6,7]. Study personnel (A.M., R.C.,

C.M.) evaluated ECGs concurrently prerace at the sympo- sium and postrace at the finish line. There were no prerace ECG abnormalities found that warranted immediate medical attention, but all subjects were informed of their ECG results at the time of ECG acquisition.

Upon completion of the half marathon, runners were instructed to go to a designated booth where a follow-up ECG was performed as soon as possible upon completion of the race. All ECGs were performed within 15 minutes of the participants’ completion. The postrace ECGs were inter- preted on-site and compared with the prerace ECGs by the same study personnel. Echocardiograms were performed on a random sampling of runners (those consenting after the race) who were judged to have abnormal or changed postrace ECGs. Two-dimensional, M-mode, and Doppler echocar- diographic studies were performed using a Sonosite M-Turbo (Bothell, Wash). Images were obtained in multiple cross-sectional planes using standard transducer positions. End-diastolic and end-systolic left ventricular (LV) cavity dimensions and anterior-ventricular (AV) septal and poste- rior free wall thickness were obtained. Resting wall motion and left atrial size were evaluated. Echocardiograms were interpreted on-site by a registered cardiac sonographer (M.S.) blinded to the clinical history or ECG pattern to assess for focal wall motion abnormalities that might require immediate intervention. The sonographer was not blinded to the fact that each patient selected to receive an echocardio- gram had an abnormal or changed postrace ECG. Echocar- diogram readings were then confirmed off-site by a cardiologist (W.A.) who was also blinded to the on-site reading, clinical history, ECG pattern, and why patients were selected to receive the echocardiogram. The final echocar- diogram interpretation of the cardiologist was used in tabulating results for data analysis. Those subjects with abnormal or changed postrace ECGs who did not have an echocardiogram performed were informed of their ECG results and referred to a primary care provider for follow-up. Prerace and postrace ECGs were then batched and analyzed by study personnel (A.M. and R.C.) and by a cardiologist (O.B.Y). At the time of this final interpretation, reviewers were blinded to all demographic data and also to which ECG was performed after the race. Electrocardiogram readings among the 3 interpreters were compared for consistency. If a discrepancy was present, the final ECG interpretation was discussed among reviewers and deter- mined by the majority opinion. Medical history and previous exercise experience were reviewed and correlated with ECGs for any trends. electrocardiogram changes were divided into the following categories of abnormalities for comparison and recorded as such: abnormalities with rate, abnormalities in rhythm, abnormalities in interval duration, and the presence or absence of ST/T-wave changes. Electrocardiogram changes also included normalization of an abnormal prerace ECG. For data analysis, “abnormalities” in interval duration were defined as results greater than 95% above or below normal values. electrocardiogram abnormalities were also

correlated with structural abnormalities seen on echocardio- graphy. Statistical analysis using the Fischer exact test was used to correlate ECG changes with specific demographic information and medical history. A 2-tailed P b .05 was considered statistically significant. Results were tabulated into an Excel spreadsheet. This study was approved by the Institutional Review Board at University of California, San Diego.

Results

Eighty-seven runners were enrolled and completed the study. None were lost from the enrolled group. There were 46 males (53%) and 41 females (47%). Age was categorized based on decade (Table 1). Patient’s ages ranged from 18 to 73 years. Most runners (60%) had previously completed four or more full or half marathons. Most runners noted that they exercised “often” or “every day.” Fifty-eight percent of runners reported no medical problems, but 9% reported a history of heart disease. Greater than half of the runners were on no medications. A family history of heart disease was present in approximately half of the runners. No runners reported chest pain before, during, or after the race or required transportation to the emergency department (ED).

Prerace ECG evaluation demonstrated 60 runners with at least 1 abnormality (69%, some subjects had more than 1 abnormality), including Sinus bradycardia (24), LV hyper- trophy (LVH) pattern (18), left atrial enlargement (LAE; 10), early repolarization (9), axis deviation from normal (10), RSR? pattern (7), first-degree AV block (5), right atrial enlargement (3), T-wave inversion (3), Right bundle branch block (2), biatrial enlargement (1), sinus arrhythmia (1), fasciular block (10), and premature atrial contractions (1). None of these were considered clinically significant. The remaining 27 subjects had normal prerace ECGs.

After the race, 54 runners (62%) had changes in their ECG (Fig. 1). Some runners had more than 1 ECG change noted. The most common change noted was atrial enlarge- ment (37 runners). Nine runners had new conduction abnormalities, 2 runners had new Q waves, and 3 runners had nonspecific ST/T-wave changes. Eight runners with ECG abnormalities before the race had resolution of these upon completion of the race. Of the runners with normal prerace ECGs, 13 (48%) developed changes in their postrace ECG. Seven had LAE, 1 had LAE and new Q waves, 3 had biatrial enlargement, 1 had RSR?, and 1 had new first-degree AV block.

There was no statistically significant difference between runners with ECG changes and runners without ECG changes when comparing sex, age, previous marathon experience, exercise, history of exercise-induced chest pain, medical history, and family history of heart disease (Table 2). For the purposes of statistical analysis, variables within categories (age, previous marathon experience, and exercise) were grouped. Multiple groupings were analyzed,

n %

Sex Male 46 53.8 Female 41 47.2 Age (y) <=20 2 2.3 21-30 17 19.5 31-40 29 33.3 41-50 17 19.5 51-60 15 17.2 61-70 4 4.6 >=70 3 3.4 Previous marathon experience None 12 13.8 1 7 8.0 2 7 8.0 3 8 9.2 >=4 53 60.1 Exercise Never 0 0.0 Rarely 0 0.0 Sometimes 2 2.3 Often 53 60.9 Everyday 32 36.8 Exercise-induced chest pain Yes 3 3.5 No 84 96.5 Medical problems Hypertension 8 9.2 Hypercholesterolemia 12 13.8 Diabetes mellitus 2 2.3 Heart disease 8 9.2 Kidney problems 0 0.0 Lung disease 8 9.2 Cancer 0 0.0 Other a 9 10.3 None 51 5.6 Medications Vitamins 8 9.2 Antidiabetic 1 1.1 Inhalers 3 3.5 Lipid-lowering agents 6 6.9 Antihypertensives 8 9.2 Anti-inflammatory 3 3.5 Anticoagulants 3 3.5 Other b 18 20.6 None 47 54.0 Family history of heart disease Yes 43 49.4 No 44 50.5

a Includes thyroid disorders and depression. b Includes medications for thyroid disorders, antidepressants, and Oral contraceptive pills.

and the P value was not impacted either way. In addition, medical problems were grouped into those that were thought to pose a risk for heart disease (group 1) and those that were less likely to do so (group 2; Table 2).

Table 1 Demographic and medical history of participants

Fig. 1 Electrocardiogram changes observed after the race.

Twenty runners with changed or abnormal postrace ECGs had postrace echocardiograms. Two of these were subjects who had normalization of their ECG postrace. Eight runners had abnormal postrace echocardiograms. The following echocardiogram abnormalities were noted: LVH (2 runners), mild diastolic dysfunction (2 runners), and LAE (1 runner). Three runners had both LVH and mild diastolic dysfunction. No focal wall motion abnormalities were noted. In 2 runners, there was some correlation between postrace ECG findings and echocardiogram abnormalities. One runner had findings consistent with LVH on both the prerace and postrace ECG. His echocardiogram demonstrated mild LVH. Another runner had findings consistent with LAE on postrace ECG, and his echocardiogram demonstrated borderline enlarge- ment of the left atrium. In the other 6 abnormal echocardio- grams, the ECGs either were normal postrace (2 runners, both with mild diastolic dysfunction) or had abnormalities that were inconsistent with what was seen on echocardiog- raphy (remaining 4 runners).

Discussion

As marathon running becomes increasingly popular, more patients may present to EDs with vigorous running- related conditions such as muscular fatigue, chest pain, and collapse [8]. Obtaining a 12-lead ECG in such situations is standard, and emergency physicians may be left to interpret a variety of ECG abnormalities in these athletes. Emergency physicians must determine whether these abnormalities represent cardiovascular disease or are benign expressions of athletic conditioning.

We have previously reported 2 cases that highlight this clinical dilemma [2]. Two runners, a 41-year-old man and a 48-year-old man, presented to the ED with chest pain and syncope, respectively, after competing in a full marathon (26.2 miles). Both runners reported no medical history. Both initial ECGs were consistent with acute myocardial infarc- tion, and both had the same ischemic pattern. Despite identical ECGs, at catheterization, one runner had a 99% occlusion of the left anterior descending artery that was amenable to stenting, whereas the other runner had no coronary artery disease [2].

From a large database of trained athletes, Pelliccia et al [3] found distinctly abnormal repolarization patterns in 1% of ECGs. A subsequent diagnosis of cardiomyopathy was made in 5 (6%) of the 81 athletes with such abnormalities who had no previous evidence of cardiac disease. Two of these 5 athletes had Major adverse cardiac events; one survived a cardiac arrest due to Hypertrophic cardiomyopathy and one died suddenly of arrhythmogenic right ventricular cardio- myopathy. Both of these athletes were previously asymp- tomatic. In contrast, cardiomyopathy or adverse cardiac events did not develop in the control group (normal ECGs). This study suggests that ECG abnormalities previously attributed to athletic conditioning may actually be expres- sions of cardiac disease.

The 12-lead ECG has been suggested as a simple and inexpensive test to strengthen the limited Diagnostic efficacy of the medical history and physical examination and is used for preparticipation cardiovascular screening for competitive athletes [9]. Electrocardiogram abnormalities in athletes can be classified as “training-related” or “training-unrelated” [10]. In general, training-related ECG changes are common and should not cause alarm. Examples of training-related

Sex					.521
Male	30	55.6	16	48.5
Female	24	44.4	17	51.5
Age (y)					.382
<=30	12	22.2	7	21.2
31-50	31	57.4	15	45.5
>=51	11	20.4	11	33.3
Previous marathon experience					.871
None/1	11	20.4	8	24.2
2-3	10	18.5	5	15.1
>=4	33	61.1	20	60.6
Exercise					.500
Sometimes/Often	35	64.8	20	60.6
Everyday	19	35.2	13	39.4
Exercise-induced chest pain					.285	(Fisher exact test)
Yes	3	5.6	0	0.0
No	51	94.4	33	100.0
Medical problems (group 1)					.401
Yes	15	27.8	12	36.4
No	39	72.2	21	63.6
Medical problems (group 2)					.678
Yes	8	14.8	6	18.2
No	46	85.2	27	81.8
Family history of heart disease					.456
Yes	25	46.3	18	54.5
No	29	53.7	15	45.5
Problem group 1: any hypertension, hypercholesterolemia, or heart disease, diabetes mellitus. Problem group 2: kidney problems, lung disease, cancer, other, or none.

ECG changes include sinus bradycardia, first-degree AV block, and early repolarization [10]. Training-unrelated changes are uncommon and may not be the result of cardiac adaptation to physical exertion. Examples include T-wave inversion, ST-segment depression, right ventricular hyper- trophy, Brugada-pattern, bundle branch blocks, and right or left atrial enlargement [10]. Further diagnostic workup is recommended in these cases to evaluate for cardiomyopa- thies or other diseases that may predispose to sudden cardiac death. In this study, atrial enlargement represented most ECG changes.

Table 2 ECG changes with respect to demographics and medical history

Postrace ECG changes (n = 54)

No.

%

No change in postrace ECG (n = 33)

No.

P

%

Abnormal ECGs pose a diagnostic question when they occur in the absence of detectable structural heart disease. In our study, the abnormalities seen on ECGs often did not correlate with echocardiogram findings. Although echocar- diograms were performed on a small subset of runners, this inconsistency raises the question of the clinical utility of the ECG in detecting structural heart disease in this setting. Most endurance-related cardiac research has focused on diagnosing cardiomyopathy in young trained athletes. Whether the same preparticipation cardiovascular screening should be applied to recreational runners is unknown. To

our knowledge, there are no studies that examine the value of prerace ECG and echocardiography in preventing long- term adverse cardiac events in recreational runners.

Limitations

Our data may not be representative of all long-distance runners, as we cannot exclude recruitment bias. Prerace echocardiograms were not performed, so we are unable to conclude if the abnormalities reported postrace represented a change from the individuals’ baseline. Although no runners reported chest pain after the race and no one required transportation to the ED, we did not obtain follow-up information. It is unknown if any runner experienced a cardiac-related event in the weeks to months after the race. In addition, interrator reliability with regard to ECG abnormal- ities was not determined. Also, postrace echocardiograms were only performed on selected patients whose ECGs were changed or abnormal. It is possible that there were runners with unchanged postrace ECGs who would have had abnormal echocardiograms. Finally, we enrolled a relatively small number of subjects.

Conclusion

Endurance sports are becoming more popular worldwide. Emergency physicians will likely encounter a greater number of adverse health effects related to these events in the future. Cardiac effects including dysrhythmias and sudden death occur with some frequency in distance runners, and the ECG is a rapid diagnostic tool for evaluation of these athletes with potential cardiac complaints. We evaluated prerace and postrace ECGs in runners of a half marathon and found that 62% had some change in their postrace ECG. We also evaluated selective runners with changes in their ECGs after the race with echocardiograms and found that 40% had abnormalities. Further study is needed to evaluate the clinical significance of these findings.

References

1. Mohlenkamp S, Schmermund A, Kroger K, et al. coronary atherosclerosis and cardiovascular risk in masters male marathon runners. Herz 2006;31:575-85.
2. Minns AB, Clark RF. electrocardiogram changes and cardiac biomarker abnormalities among two marathon runners. J Emerg Med 2010;38(2):159-61 [Epub 2008 Dec 11].
3. Pelliccia A, Di Paolo FM, Guantrini FM, et al. Outcomes in athletes with marked ECG Repolarization abnormalities. N Engl J Med 2008; 358(2):152-61.
4. Maron BJ, Pelliccia A. The Heart of trained athletes: cardiac remodeling and the risks of sports, including sudden death. Circulation 2006;114:1633-44.
5. Pelliccia A, Maron BJ, Culasso F, Di Paolol FM, et al. Clinical significance of abnormal electrocardiographic patterns in trained athletes. Circulation 2000;102:278-84.
6. Friedman HH. Diagnostic electrocardiography and vectocardiography. New York: McGraw-Hill; 1971.
7. Sokolow M, Friedlander RD. The normal precordial and limb lead electrocardiogram. Am Heart J 1949;38:665.
8. Siegel AJ, Januzzi J, Sluss P, Lee-Lewandrowski E, Wood M, Shirey T, et al. Cardiac biomarkers, electrolytes, and other analytes in collapsed marathon runners. Am J Clin Pathol 2008;128:948-51.
9. Corrado D, Pelliccia A, Bjornstad HH, et al. Cardiovascular pre- participation screening of young competitive athletes for prevention of sudden death: proposal for a common European protocol. Eur Heart J 2005;26:516-24.
10. Corrado D, Biffi A, Basso C, Pelliccia A, Thiene G. 12-lead ECG in the athlete: physiological versus pathological abnormalities. Br J Sports Med 2009;43:669-76.