Word count: 4367 - Sportsci



Word count: 4367

Electrocardiography In Elite Athletes

F. CARRE and J.C. CHIGNON*

Department of Physiology, Pontchaillou Hospital

35003 Rennes, France

*National Institute of Sports, 11 Avenue du Tremblay,

75012 Paris, France.

Electrocardiographic (ECG) features commonly observed in top ranking sportsmen were first described in 1929 by Hoogerwerf. The development of new non-invasive exploration methods (i.e. cardiac echocardiography and magnetic resonance imaging) in the early 1970s offered physiologists the means of investigating the so-called "Athlete's Heart Syndrome" recognized by its specific ECG features.

Standard ECG tracings have the advantage of low cost, ready availability and ease of use and remain an essential investigation tool. In trained subjects, ambulatory ECG monitoring and stress testing ECG are particularly important as they provide supplementary information to the classical 12-ECG which records only a brief period of cardiac electrical activity.

The features of the athlete's ECG basically reflect the heart's normal physiological adaptation to repetitive physical training. However several unusual patterns appear to be quite similar to pathological aspects occurring in different heart diseases. It is thus essential to acquire a full understanding of the ECG patterns in the elite athlete.

Interpretation of the athlete's ECG

A highly trained athlete is usually defined as a subject who practices at least ten hours a week at a level of intensity reaching at least 60 percent of his maximal oxygen consumption ([pic]O2max). Consequently, and athlete's ECG must always be interpreted in light of his individual level of training, both qualitatively and quantitatively, and in accordance with the physical examination, functional signs and his personal and familial cardiovascular risk factors, including age.

The most common ECG features described in elite athletes can be observed in all age groups and in both men and women, however they are not always found in every elite athlete. They result from physiological adaptations to physical conditioning and should not be immediately interpreted as markers of heart disease. The mechanisms underlying the disturbances observed on the athlete's ECG are not yet fully understood although modifications in autonomic nervous system tone and cardiac hypertrophy are often proposed as significant explanations.

Modifications in autonomous nervous system tone have been described in the athlete's heart syndrome on the basis of biochemical and pharmacological tests. Another way of evaluating the effect of changes in parasympathetic and sympathetic tone is to study heart rate variability. Characteristically, there is an increase in parasympathetic tone and a decrease in sympathetic tone (perhaps through the effect of lower baroreceptor sensitivity). At rest, vagotonia appears to predominate whereas during exercise the deceased sympathetic drive results in the slower heart rate observed in athletes compared with untrained subjects performing the same work load.

Cardiac hypertrophy in the athlete was first suspected by Henschen in 1899 on the basis of chest percussion and has been confirmed by non-invasive morphology investigations including radiology, echocardiography and more recently magnetic resonance imaging. It is described as a four-chamber harmonious wall hypertrophy-chamber dilation which can be observed at all ages and appears to be totally reversible after deconditioning.

There is some controversy in the literature as to the real incidence of ECG disturbances. For example, in two studies based on a large sample population (Venerando published a series of 12,000 subjects and in our own personal unpublished work we investigated 6,487 subjects) the global prevalence of ECG disturbances was found to be 13 and 44 percent respectively. This difference could be explained by differences in methodology in the training level since the ECG criteria for diagnosis of cardiac hypertrophy have not been standardized.

The mean ± SD values of classical ECG criteria as observed in our study are given in Table I in comparison with the ranges classically described in a standard population. In general, the ECGs of athletes lie within standard limits. A few trends which increase with training level can however be seen. The durations of the PR interval, the QRS complex and the corrected QT interval increase with the Sokolow-Lyon Index and frontal QRS axis turns to the left. Even though the ECGs of elite athletes lie within normal limits, different patterns of ECG disturbances have been described.

For the purposes of this review, we have divided these changes into rhythm disorders, atrio-ventricular conduction impairment, cardiac hypertrophy related ECG criteria and disturbed repolarization. Finally, we shall try to specify the potential differences observed in endurance versus resistance in the trained athlete.

Changes in cardiac rhythm

Hypokinetic Arrhythmias

The respective incidences of changes in cardiac rhythm and hypokinetic arrhythmias are summarized in Table II.

Resting sinus bradycardia is the most common finding among trained athletes. It is difficult to determine the real incidence of athlete's bradycardia due to the lack of a common definition of bradycardia. The incidence varies from 8 to 85% in studies using the cut-off of 60 beats per minute, and in our study, we found only 9% of our athletes with a resting heart rate below 50 beats per minute. Controlled Holter recordings have shown a significantly lower mean hourly heart rate. Training undoubtedly affects the incidence of bradycardia but the role of individual sensitivity and the mechanisms of training-induced bradycardia have yet to be established. Classically, the alterations in the autonomous nervous system described above would have an effect, but some studies have shown that lower intrinsic heart rate is also related to athlete's bradycardia. In most cases, the bradycardia is benign as confirmed by normal rhythms recorded during stress testing and also by the persistence of physiological circadian variations (lower nocturnal heart rate) on Holter recordings. Rarely, the bradycardia is associated with dizziness, syncope or hyperkinetic arrhythmias due to vagal tone. In general, these symptoms disappear with deconditioning. Electrical stimulation is rarely needed and usually concerns older athletes in whom a latent sinus node disease is unmasked by the increased vagal tone. The resting heart rate in individuals with athlete's bradycardia correlates with their individual level of peak training, and is used as a criteria for evaluating their level of training although it is not well correlated with performance or [pic]O2max. A better index of training level would be the heart rate recovery curve. The rapidity at which the heart rate returns to the basal level (or near basal level) would be an indication of a good level of training. An unusual disturbance of the resting sinus rate which cannot be explained by a change in the training regimen, is commonly considered to be a feature of overtraining.

Other hypokinetic dysrhythmias also concern the sinus rhythm and are related to altered autonomous tone. They disappear during stress training. These modifications are frequently observed on ambulatory ECG recordings, particularly at night. They are of no prognostic significance.

The prevalence of sinus dysrhythmia , the so-called "respiratory arrhythmia", would appear to be significantly higher in athletes than in the standard population, but in fact the apparent sinus dysrhythmia disappears when the variability of R-R interval as a function of basal heart rate is taken into account (R-R interval variation increases with decreasing heart rate). This is a well-recognized ECG pattern on ambulatory ECGs where the sinus pauses during both awake and (especially) sleeping hours are significantly longer on athlete's recordings than on control recordings.

Ectopic atrial rhythm including the wandering atrial pacemaker or coronary sinus rhythm have also been described.

Nodal rhythm is more frequent in elite athletes. The escape threshold varies from 45 to 65 beats per minute and in some cases (in 15% of the subjects in our study) escape rhythm totally disappears only above 100-120 beats per minute.

Idioventricular rhythm (Figure 1) is the event of a low sinus rate and/or of sinus pauses. This cardiac rhythm originates in pacemaker cells at a rate of 40 to 100 beats per minute.

Hyperkinetic Arrhythmia

Hyperkinetic arrhythmia involves premature supraventricular and ventricular beats. These disorders can be detected on the resting ECG but most of the studies have used Holter monitoring to best quantify these episodes of arrhythmia. A training session during the monitoring period is useful because ECG stress training does not always produce significant episodes. It is important to study arrhythmia during exercise and during recovery to clarify the links with autonomous tone and the epinephrine effect.

Supraventricular Arrhythmias

The incidence of premature supraventricular beats observed in trained athletes (37.1 to 100%) is similar to, or higher than, that seen in the standard population (20-80%). Some authors relate premature supraventricular arrhythmias to training level and suggest athlete's bradycardia could be an explanation. Most often, these premature beats are isolated and infrequent (less than 15 to 20 per 24 hours). They are asymptomatic and may disappear during exercise. Complex supraventricular tachyarrhythmias which provoke palpitations are rarely described (0.5 to 5%) and suggest an underlying heart disease such as the Wolff-Parkinson-White syndrome or prolapsus of the mitral valve. The role of vegetative imbalance has been suggested in cases of paroxysmal atrial fibrillation.

Ventricular arrhythmias

On resting ECG recordings, the incidence of ventricular arrhythmias in trained athletes is similar to or much higher (0.5 to 4%) than in the standard population (0.6 to 0.7%). On the basis of Holter studies, most authors conclude that the incidence in athletes (30 to 45%) is the same as in untrained subjects (16-55%) but in one controlled study, the incidence was higher in athletes (70%). Generally, premature ventricular beats are unifocal, isolated, infrequent (less than 50 per 24 hours), asymptomatic and disappear at the onset of exercise.

In our clinical experience with regularly screened athletes, we distinguished (Figure 2) between old asymptomatic arrhythmia, which disappears during exercise and reappears during the slow phase of recover and which we consider to be benign, and a newly occurring, often symptomatic (unexplained decline in performance) ventricular arrhythmia which usually persists or becomes worse during stress testing. This situation, which suggests catecholamine sensitive focal arrhythmia, always requires a complete cardiac examination to eliminate a latent heart disease. Overtraining, which sometimes provokes hyperkinetic arrhythmias through changes in biological mechanisms, must be suspected only if the cardiac examination is normal.

Other complex ventricular arrhythmias such as multifocal or repetitive premature ventricular beats, ventricular tachycardia and R on T phenomena appear to have the same incidence in trained and untrained individuals. Some authors have observed paroxysmal ventricular tachycardia in 0 to 7.5% of athletes (ventricular tachycardia is classically nocturnal but sometimes appears in daytime) and in 0 to 5.7% in untrained subjects. Here again some authors describe a higher incidence of complex arrhythmias in trained subjects and explain their controversial results by the training level in the general population. For these authors regular moderate physical training could protect against ventricular arrhythmias while very intensive training could favor them, perhaps through a prolonged QT interval. In contrast with cases of pathological cardiac hypertrophy, no study has been able to demonstrate a correlation between ECG or echocardiographic cardiac hypertrophy and hyperkinetic arrhythmia in athletes.

In concluding this chapter, it can be stated that hypokinetic arrhythmias in athletes are common and benign. The discovery of an episode of hyperkinetic arrhythmia, particularly ventricular hyperkinetic arrhythmia, in an athlete often raises the question as to how many single extra beats should be tolerated. This is especially true in high level trained subjects who undertake maximal exercise regularly and often encounter the well-known adrenergic stress. It would appear that the prevalence of hyperkinetic arrhythmia is nearly the same in trained and untrained people and that cardiac adaptation to intensive training itself is not a determinant cause of malignant arrhythmia. Therefore the discovery of a recent or serious episode of hyperkinetic arrhythmia in an elite athlete would require a full cardiac examination with Holter monitoring, stress testing, echocardiography, and if necessary an electrophysiologic study.

Impaired atrio-ventricular conduction

First or second (with a Luciani-Wendkeback period, see Figure 3) degree atrio-ventricular block is relatively common in athletes (Table III). Inversely, third degree functional block is rarely described and until now the Mobitz type II and higher degree atrio-ventricular blocks must be considered as pathological and require cardiologic screening. These disorders result mainly from changes in autonomous tone and disappear during stress testing and or pharmacological tests.

Their higher, although intermittent (nocturnal predominance), incidence in Holter studies (Table III) would confirm their functional character. A correlation with training intensity has been reported. Although the same physiological explanations have been proposed as for hypokinetic arrhythmia and conduction disorders in athletes, it must be noted that no real correlation has yet been described linking the two phenomena.

The prevalence of the pre-excitation syndrome (i.e. the Wolff-Parkinson-White syndrome and the short PR syndrome) in athletes is nearly the same as in the standard population (0.16 to 1%) even though changes in autonomous tone could unmask accessory pathways. The discovery of a pre-excitation syndrome in trained subjects always requires a full cardiac exploration.

Ecg Disturbances Partly Related To Cardiac Hypertrophy

As noted above, cardiac hypertrophy in the athlete is described as a classical "physiological" example of adaptative increase in heart volume. Many different ECG criteria have been proposed for the diagnosis of athlete's cardiac hypertrophy based on isolated voltage criteria (i.e. the Sokolow-Lyon Index) or voltage and non-voltage criteria (i.e. the Romhilt-Estes Point Score System) such as intraventricular conduction delays and/or repolarization disturbances. Unfortunately, the different ECG parameters of cardiac hypertrophy observed in athletes are poorly correlated with the results of non-invasive investigations such as echocardiography or with those of invasive or anatomic studies. This can be explained, at least partially, by the fact that these populations are comprised of young and physically fit individuals. Thus the ECG does not appear to be an extremely useful tool for the assessment of cardiac hypertrophy in the athlete.

Nevertheless, it is essential to recognize the features of elite athletes' ECGs. Increased P wave amplitude, with or without notching, can be observed although several studies failed to demonstrate any significant difference compared with matched controls. Right ventricular hypertrophy, based on the classical Sokolow-Lyon Index (RV1 + SV5), has been reported in 4.5 to 6.9% of athletes.

In heterogeneous and small samples of athletes, the incidence of left ventricular hypertrophy based on a Sokolow-Lyon Index (SV1 + RV5 or RV6) > 35 mm has been reported to vary from 8 to 85% compared with 5% in the general population. Inversely, in our study of a large population of trained subjects (Table I), and in the study by Venerando et al. (12,000 subjects), there was no real enhancement of the Sokolow-Lyon Index.

The use of new cardiac hypertrophy ECG criteria, including total QRS amplitude in 12-lead ECGs, appears to be helpful and in our own study (Table IV) we found that this sum (mean: 192 ± 40 mm) was higher than the classical sedentary sum (< 128 mm) but clearly less than the sums described in heart diseases (aortic stenosis > 244 mm; aortic regurgitation > 246 mm). Based on vectocardiographic criteria, the prevalence of left ventricular hypertrophy is about 40% (37-46%).

Electrical wave delays have also been studied in athletes. The incidence of right and left atrial hypertrophy is low. The duration of QRS complexes is correlated with the size of the heart chambers and many authors suggest that the best criteria for right ventricular hypertrophy in the athlete is the presence of intraventricular conduction delay. This delay, which appears on the ECG tracing as a notching or slurring of the QRS complex on D3, aVF and on the right precordial leads, is often observed (3.2 to 70%). These features suggest, as does the well-known incomplete right bundle branch block (prevalence 1.7 to 51%), an asymmetrical cardiac hypertrophy with right ventricular predominance. Though vectrocardiographic studies have also noted a high frequency of right ventricular hypertrophy (18 to 30%) this explanation is questionable since echocardiographic data do not offer a confirmation. Incomplete right bundle branch block does not appear to be linked to changes in autonomous tone since it persists during stress testing. Ventricular apical thickness may be involved. Complete right bundle branch block is much rarer (0.08 to 0.31%) and left bundle branch block is normally not observed in the elite athlete.

Unlike cardiac patients, and in spite of these cardiac hypertrophy ECG criteria, the QRS axis is often normal. A vertical QRS axis may be observed (10 to 27%) and left deviation is seldom reported (10 to 12%). Similarly, associated pathological repolarization is not common.

In summary, trained athletes show a high incidence of cardiac hypertrophy based on ECG criteria. These phenomena, including right bundle branch block, are related to physical training since the incidence decreases significantly with deconditioning. Nevertheless, these features cannot be fully explained by cardiac hypertrophy alone. Besides anatomic heart adaptation, other factors including age, body weight, body surface area, fat-free weight and depth of the heart in the chest may also play a role. ECG criteria of cardiac hypertrophy are however, as are echocardiographic features, quite different in elite athletes as compared with those described in the patients with heart diseases.

Repolarization disturbances

Repolarization disturbances are a striking feature observed in "athlete's heart syndrome". These phenomena lie between a physiologic and pathologic state (i.e. pericarditis, ischemia, metabolic disturbances.…). It is difficult to give a precise assessment of their prevalence partly because of seasonal and career variations. Holter monitoring is less useful than stress testing in this situation. No single explanation has been proposed for these disturbances, although changes in autonomous tone and/or cardiac hypertrophy and/or electrolyte abnormalities have been proposed. These repolarization disturbances are generally asymptomatic.

Several classifications have been proposed. We think the most useful is the descriptive classification developed by Zeppilli and Caselli. These authors propose four criteria. Criteria (a) and (b) are classically described asminor repolarization abnormalities.

Criteria (a), the so-called "early repolarization syndrome" is the most frequent (10-100%). The top of the ST-T segment elevation often has a dip in the initial portion. It has been speculated that changes in autonomous tone could be the cause. Sympathetic tone decrease reveals inherent a non-homogeneity phase of the ventricular repolarization, the epicardium repolarizing first. The ECG pattern, well-correlated with duration and training level is age-dependent and benign. This is supported by the fact that it disappears either at the onset or early during stress testing.

Criteria (b) is classically characterized as negative T waves in inferior (D2, D3 or aVF) or right precordial (V1-V3) leads; low amplitude or flat T waves can also be observed. Described in 3-31% of the trained population, they regress as a general rule during exercise. They must be related to vagotonic-induced heterogeneity of the myocardial action potential. They are sometimes associated with echocardiographic criteria for cardiac hypertrophy.

Criteria (c) and (d) are described as marked repolarization disturbances. In our experience as in that of Venerando, the prevalence is relatively low (0.6-2.8%). A complete cardiac work-up is always needed.

Criteria (c) is defined as JT segment depression with positive low-voltage isoelectric or diphasic T waves. This feature which evokes subepicardial ischemia is a questionable physiological adaptation and must be assessed carefully because it disappears inconsistently during stress testing or after a long period of deconditioning.

Criteria (d) is defined as T wave inversion in the left precordial leads (V4 - V6) which also disappears inconsistently during stress testing (Figure 4).

In a study involving 98 athletes who presented features (b), (c) and (d), Zeppilli et al. reported no demonstrable heart disease in 53%, prolapsus of the mitral valve in 37%, hyperkinetic heart syndrome in 3% and hypertrophic cardiomyopathy in 4%. More recently certain authors have stressed that negative T waves on the right precordial leads in athletes, especially when associated with incomplete right bundle branch block or premature ventricular beats with a left bundle branch block configuration, may reveal right ventricular dysplasia.

Other repolarization disturbances have been described in the elite athlete including the common and benign evident U wave (especially in precordial leads) and a prolonged corrected QT interval (prevalence 10 to 15%) which could be explained by changes in autonomous tone and for which, in trained subjects, no real relationship with ventricular arrhythmias has been observed.

Thus the prevalence of ECG and vectocardiography patterns of repolarization disturbances, especially minor abnormalities, is higher in trained individuals than in the untrained population. No unequivical explanation has been proposed. These features vary spontaneously and are not correlated with physical fitness. Their interpretation must take into account different factors including age, ethnic origin, training level and symptoms. Venerando has stressed the criteria of benign disturbances: healthy and totally asymptomatic athletes with good physical capacity (VO2max), normal duration of QRS complex and lack of (or constantly reversible) spontaneous (exercise) or induced (pharmacodynamic tests) ECG abnormalities.

In the present state of the art, the recent discovery of marked repolarization abnormalities requires a compete cardiac work-up, including at least stress testing and echocardiography.

Comparison Of "Endurance" And "Power"

Physiological adaptation is generally divided into two categories resulting from the effects of two types of training methods: aerobic and anaerobic. Actually, the results of both ECG and echocardiographic studies are rather controversial. This can be explained, at least in part, by the fact that most athletes undertake both types of training simultaneously.

In our personal study (Figure 5) we found that the prevalence of bradycardia and incomplete right bundle branch block was higher in endurance than in power athletes. Inversely, premature ventricular beats occurred more frequently in power athletes. Some authors stress the fact that sinus pauses longer than 2,000 ms, ECG criteria of left ventricular hypertrophy and prolongation of the corrected QT interval are more frequent in endurance athletes. On the other hand, some authors suggest that marked repolarization disturbances tend to be associated more readily with isometric training.

Suggested readings:

1- Carré F. and J.C. Chignon. Advantages of electrocardiographic monitoring in top level athletes. Int. J. Sports Med. 12: 236-240, 1991

2- Ferst J.A. and B.R. Chaitman. The electrocardiogram of the athlete. Sports Med. 1: 390-403, 1984

3- George K.P., L.A. Wolfe, G.W. Burgraff. The "Athletic Heart Syndrome". Sports Med. 11: 300-331, 1991

4- Huston T.P., J.C. Puffer, W.M. Mc Millan-Rodney. The athletic heart syndrome. New Eng. J. Med. 313: 24-32, 1985

5- Lichtman J., R.A. O'Rourke, A. Klein et al. E.C.G of the athlete. Arch. Intern. Med. 1323: 763-770, 1973

6- Rost R.and W Hollmann. Athlete's heart- a review of its historical assessment and new aspects. Int. J. Sports Med. 4: 147-165, 1983

7- Venerando A. Electrocardiography in sports medicine. J. Sports Med and Phys. Fitness. 19: 107-128, 1979

8- Zeppilli P., A. Pelllicia, M.M Pirrami et al. Ethiopathogenetic and clinical spectrum of ventricular repolarization disturbances in athletes. J. Sports Cardiol. 1: 41-51, 1984

Figure Legends

Figure 1. Intermittent idioventricular rhythm in a long distance runner. Precordial lead V3, amplitude divided by two.

Figure 2. Two typical cases of isolated premature ventricular beats observed on 24-hour Holter recordings in athletes.

Subject 1 was a soccer player with old, asymptomatic, isolated premature ventricular beats. Hourly frequency of premature beats does not vary.

Subject 2 was a weight lifter with recent, symptomatic, isolated premature ventricular beats. A peak frequency occurred during two training sessions (T)

h = hours of monitoring, nb•h-1 = number of premature ventricular beats per hour.

Figure 3. Asymptomatic second degree atrio-ventricular block with a Luciani-Wenckeback period observed in a cyclist during the competition period. P waves are noted with an arrow (—).

Figure 4. An asymptomatic, 35-year-old, well-trained long distance runner.

Resting ECG shows incomplete right bundle branch block and a negative T wave in the V5 lead.

Maximal exercise ECG shows a significant (2 mm) JT depression (V5).

Recovery ECG (5 min) showing a normalization of the T wave on V5.

Exercise thallium myocardial scintigraphy was normal and echocardiography showed an asymmetrical septal hypertrophy (12 mm).

Figure 5. Respective prevalence of resting ECG features observed in endurance athletes (n = 5,700) and power athletes ( n = 526) in our own study. BRA = sinus bradycardia, AVB = atrio-ventricular block, RBBI = incomplete right bundle branch block, SVPB = premature supraventricular beats, VPB = premature ventricular beats.

Table Legends

Table I. Classical ECG criteria: comparison between trained individuals and general population.

These data were obtained in men and women 15 to 40 years of age.

Data concerning trained individuals were observed in athletes examined at the French National Institute of Sports. Three training level groups were described: I (three to five hours per week), II (five to ten hours per week), III (more than ten hours per week, national team level).

Data concerning the general population were described by Blondeau and Hiltgen, 1980 (15-19 years of age, n = 200; 20-29 years, n = 200; 29-39 years, n = 200).

* T wave amplitude measured on the precordial lead V5

** QT corrected for a heart rate of 60 beats per minute

*** QT corrected using the Bayes formula.

Table II. Incidence (%) of athlete's hypokinetic arrhythmia

Ranges are based on the highest and lowest values reported in the literature and were observed in controlled and uncontrolled studies.

bpm = beats per minute

(--) = no data available.

Table III. Incidence of athlete's atrio-ventricular conduction impairment

Ranges are based on the highest and lowest values reported in the literature and were observed in controlled and uncontrolled studies.

* One study (Viitasalo et al., 1982) reported an incidence of 8.6% for Mobitz II atrio-ventricular blocks which were, very probably, in fact Luciani-Wenckebach type II atrio-ventricular blocks with a very small increment in the PR interval duration (personal communication of the authors).

Table IV. Comparison of two ECG criteria (mean ± SD) for cardiac hypertrophy in trained subjects, (n = 730).

Three training level groups were described (see Table I).

[pic]O2max = maximal oxygen consumption.

S-L Index = Sokolow-Lyon Index (SV1 + RV5 or RV6).

Total QRS = sum of the QRS complex amplitudes in the twelve ECG leads.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download