The Athletic Heart Syndrome
Ruling Out Cardiac Pathologies
James C. Puffer, MD
Exercise and Sports Cardiology Series
THE PHYSICIAN AND SPORTSMEDICINE - VOL 30 - NO. 7 - JULY 2002
In Brief: Patients who participate in regular vigorous or strenuous physical activities undergo significant changes in cardiac structure and function. Occasionally, these changes may be confused with those of hypertrophic cardiomyopathy (HCM). Differentiating between athletic heart syndrome and HCM requires careful examination. ECG and echocardiograms may be helpful, but other techniques such as detraining can also be useful in resolving the issue. Detraining produces regression of cardiac features in patients with athletic heart syndrome, while enlarged cardiac features remain unchanged in those with HCM.
The heart undergoes profound changes in response to athletic training, producing in well-conditioned athletes morphologic, functional, and electrophysiologic alterations collectively known as the athletic heart syndrome. But a perplexing situation for the clinician is the athlete whose problem cannot be attributed to the athletic heart syndrome. The usual clinical situation is one in which exercise-associated collapse, near-syncope, or syncope is the presenting complaint, and a carefully taken history and physical examination reveal little. Further workup will usually include an electrocardiogram (ECG) and/or an echocardiogram. If the ECG demonstrates abnormal findings or the echocardiogram reveals septal or wall measurements that fall within the diagnostic gray zone, the dilemma becomes whether these findings represent underlying pathology or the athletic heart syndrome.
Although the significant role that an athlete's heart plays in the ability to perform maximum exertion was recognized years ago,1,2 vigorous athletic activity was thought to be deleterious3 until work by Deutsch and Kauf4 dispelled this notion. Their work raised several questions about heart size and athletic ability; among them: Are championship athletes naturally endowed with large hearts, or do the changes in cardiac dimension occur from vigorous physical training?
Changes in Heart Morphology
Changes in cardiac dimension bring about significant mechanical advantage when cardiovascular hemodynamics are considered. The athlete who needs a high capacity for oxygen transport benefits from a large stroke volume, a low heart rate, and a hypertrophied ventricular wall.5,6 While the heart continues to function adequately as a pump by altering its rate and contractility when confronted with sudden demand, chronic demand causes dilation and hypertrophy.
Morganroth et al7 used echocardiography to assess left ventricular (LV) dimensions in 56 athletes and demonstrated that LV end-diastolic volume and mass increased in isotonic athletes (chronic volume demand) compared with controls. Isometric athletes (chronic pressure demand), in contrast, had increased LV mass but normal end-diastolic volume. On average, wall thickness was greater in the isometric athletes.
Such changes in heart dimensions, in fact, are responsible for performance. Turpeinen et al8 used magnetic resonance imaging and echocardiography to assess cardiac dimensions in nine male endurance athletes and eight sedentary controls. Their study demonstrated that the trained subjects manifested significant increases in LV mass and increased long-axis LV diameter. Most important, maximum oxygen consumption significantly correlated with long-axis LV diameter.
Large heart or enlarged heart? Ehsani et al9 answered the critical question about a natural or exercise-induced origin for large hearts. Their longitudinal study assessed cardiac dimension in response to training and detraining in competitive swimmers over a 12-week period. During the 9-week training session, LV end-diastolic diameter increased significantly (48.7 mm at baseline vs 52 mm posttraining). Correspondingly, LV free-wall thickness also increased significantly (9.4 mm at baseline vs 10.1 mm posttraining). After 3 weeks of detraining, echocardiographic measurements revealed significant regression in all the athletes.
Wall-thickness changes in response to exercise are highly variable. Pelliccia et al10 used echocardiography to measure LV dimensions in 947 elite male athletes. Wall thickness varied from less than 7 mm to as much as 16 mm (figure 1). Fifteen of the athletes with the greatest wall thicknesses were rowers or canoeists. Pelliccia et al11 found similar results among 600 women athletes, although the range of values was more narrow.
Considerable evidence supports the notion that the changes in cardiac dimension in athletes represent normal adaptive responses to strenuous exercise. Using echocardiography, Sugishita et al12 assessed cardiac morphology in 31 runners and 17 judo athletes and compared this data with that obtained from normal controls and patients with various cardiac conditions. They found that the ratio of LV radius to wall thickness was normal in runners but increased in patients with aortic regurgitation and dilated cardiomyopathy. Therefore, the changes in the long-distance runners' hearts could clearly be distinguished from those in patients with volume overload pathology.
Assessing Functional Changes
The advent of positron emission tomography (PET) has provided new noninvasive technology for assessing the heart. Nuutila et al13 used PET to assess metabolic function in seven male endurance athletes and seven sedentary controls. Although glucose uptake was enhanced in the whole body and skeletal muscle in the athletes when compared with controls, myocardial glucose uptake was reduced, suggesting that the athletes' hearts either had altered energy requirements or had used a different substrate for demands. Turpeinen et al8 showed that fatty acid oxidation was not used preferentially as an alternative energy source.
While these data suggest unique energy requirements in the athlete's heart, no published studies have addressed coronary perfusion reserve in the athletic heart. A pilot study has sought to answer this question using 13N ammonia and PET before and after dipyridamole administration in competitive athletes and age-matched controls (Puffer and Larson, 1998, unpublished data). Data suggest that regional myocardial blood flow is similar in both groups and that the coronary reserve is normal to increased after dipyridamole administration. No relationship has been demonstrated between LV wall thickness and blood flow or coronary reserve.
Several common ECG findings have been documented in athletes. These changes include alterations in rhythm, conduction, repolarization, and precordial voltage. Most of these effects derive from an increase in vagal tone or an alteration in the total neural input to the heart resulting in downregulation in sympathetic drive. Alternatively, denervation studies either done surgically or induced chemically with atropine and propranolol have shown that trained subjects consistently manifest lower intrinsic heart rates than sedentary controls, whether at rest or during exercise. This finding suggests an intrinsic cardiac component to these changes.14 The effect of training on both autonomic input and intrinsic pacemaker adaptation most likely accounts for most of the ECG changes described below.
Bradycardia and arrythmia. Resting sinus bradycardia is the most frequent finding in ECGs of well-conditioned athletes (table 1); more than half of all athletes engaged in dynamic activities have this finding. Resting rates can be as low as 25/min.15 Sinus arrhythmia occurs almost as frequently in athletes and is significantly more common than in the general population.14 Other rhythms have been noted less frequently but also occur more often in athletes than in sedentary controls. All of these rhythms extinguish with exercise as sympathetic drive is increased.
Abnormal conduction and blocks. Conduction abnormalities also occur commonly in well-conditioned athletes. First-degree atrioventricular block occurs in 6% to 33% of these athletes, significantly higher than the 0.65% found in the general population.14 Interestingly, athletes in whom first-degree heart block is absent often have relative prolongation of the PR interval.16 As expected, the PR interval normalizes or shortens following exercise.
Second-degree atrioventricular block has also been reported in well-conditioned athletes. While the frequency of Möbitz type 1 and 2 heart block in an asymptomatic population is less than 0.003%, Möbitz type 1 block has been reported in 0.125% to 10% of athletes.14 Zeppilli et al17 investigated 10 athletes with Möbitz type 1 second-degree heart block. They found that a Valsalva maneuver normalized conduction in 7, exertion resulted in normal conduction in 9, and administration of atropine corrected the abnormality in all 10. Complete remission occurred after training cessation. Several athletes demonstrated variation in the degree of block on previous ECGs, which was primarily related to training intensity. Five additional cases, as well as two cases of third-degree heart block, were reported in a study of 12,000 athletes.18 Sympathetic inducing maneuvers normalized the conduction in these athletes, and, as in the other cases, sinus rhythm was restored when training stopped. Nine years of follow-up did not reveal evidence of progression of heart block.
Although it is present in only 0.06% of the general population, junctional rhythm has been reported in 0.31% to 7% of athletes.14 Postexercise junctional rhythm has been observed in well-trained athletes as well. While a complete bundle branch block has been described in a few athletes, no reliable evidence suggests that it occurs more frequently in athletes or is induced by training.
Athlete ECGs. Studies in which comparisons of athletes and nonathletes are derived from routine 12-lead ECG tracings are somewhat problematic, because only a few seconds of cardiac electrical activity are recorded on these ECGs. Depending on the relative balance between sympathetic and parasympathetic activity at the moment of measurement, many abnormalities may be missed. This problem can be avoided with ambulatory ECG technology, which captures electrical activity over a prolonged time. Hanne-Paparo and Kellerman19 demonstrated that athletes had consistently lower heart rates, higher incidence of first- and second-degree block, and more frequent and longer sinus pauses than controls. The incidence of ventricular premature beats was similar. While two controls had short runs of ventricular tachycardia, it was absent in athletes. Subsequent studies have revealed similar findings.
Repolarization. Alterations in repolarization also occur frequently in athletes and can be categorized into four distinct patterns14:
By far the most common ST-segment and T-wave changes seen in athletes are those of early repolarization. Cross-sectional studies reveal frequencies of this phenomenon that range from 10% to 100%.14 In nonathletes, the physiologic basis for this phenomenon is thought to be nonhomogeneous repolarization of the ventricle with the epicardium repolarizing first. Early repolarization is more common in African-Americans and is usually most prominent in leads V2 to V4, with reciprocal changes in the inferior leads. J-point elevation is usually present, and occasionally T-wave inversion or tall, peaked T waves can be seen.
According to one theory, the increased frequency of early repolarization seen in competitive athletes has been related to a training-induced decrease in resting sympathetic tone, which then uncovers an inherent asymmetry of repolarization in those so inclined.20 Although this theory is consistent with normalization in exercise and disappearance in detraining, normalization does not always occur after deconditioning, so the theory must still be considered speculative. Furthermore, ST-segment normalization with exercise can occur in cardiomyopathies, and therefore exercise-induced normalization of ST-segment elevation does not necessarily rule out an underlying pathologic basis.
ST-segment depression is found much less frequently than the changes of early repolarization. One large review21 found a 0.1-mV depression in 3% of bicyclists studied. With exertion, the ST segment usually returns to baseline, except for findings in one small study.22 However, since pretraining ECGs were not available, these changes could not be shown to be the result of physical training.
T waves. Alteration in the T wave can occur in one of two ways: tall and peaked, or inverted. Tall, peaked T waves are frequently seen as part of the early repolarization syndrome. Unfortunately, the literature does not document whether tall, peaked T waves can occur in the absence of ST-segment changes, although this phenomenon has been seen by physicians. T-wave amplitude is not related to hyperkalemia, since the amplitude of the T wave decreases as serum potassium rises after exertion. Amplitude does rise in the precordial leads as training progresses, but this occurs with ST-segment elevation as well. Tall, peaked T waves occur concomitantly with physical training, but they are clearly not an isolated finding. Increased amplitude may reflect repolarization changes or may be merely secondary to repolarization of increased ventricular mass.
Frank T-wave inversion occurring across the precordium and/or in the limb leads has been well documented in cross-sectional studies. Normalization occurs following maximal exertion or other sympathetic maneuvers, and inversions disappear with cessation of training.23 A comparison of athletes with and without T-wave inversion has been reported.24 All of the athletes had normal thallium treadmill tests. Those with inverted T waves were shown to have impressive increases in precordial QRS voltage, interventricular septal thickness, LV posterior wall thickness, and calculated LV mass. Some researchers25 have suggested that T-wave inversion results from differences in action potential duration of myocardial cells. Increasing the sympathetic stimulus could normalize the T wave by making the action potential duration more uniform, which would explain why T-wave inversion occurs and why it can normalize with exertion and then disappear with deconditioning.
A variation of frank T-wave inversion is the presence of biphasic or terminal negativity, typically occuring in leads V3 to V5. Whether this is a training-induced effect or simply a part of a "juvenile pattern" is unclear. While no evidence exists that the frank or terminal T-wave inversion arising from training is an indicator of pathology, one must exercise care in evaluating deep T-wave inversions, which are accompanied by symmetric T-wave contour, ST depression, prolonged QT interval, or absence of normal septal Q waves. These patterns are more likely to indicate significant underlying pathology.
Voltage changes. Voltage changes in the QRS complex are commonly seen on ECGs of athletes. Since the diagnostic criteria for right and left ventricular hypertrophy (LVH) are not standardized across studies, comparison of prevalence in athletes and nonathletes is difficult. Voltage criteria for LVH in athletes varies from 8% to 76%.26 The deflections of SV1 and RV5 average more than 35 mm in five of six studies;14 in the young they may be more than 45 to 53 mm. Many authors have used this as a method for assessing LVH on the ECG, although it is known that this criterion may be falsely positive in young adults. These studies also show that dynamic athletes demonstrate consistently greater voltages than athletes who participate in static activities such as weight lifting.
Similarly, four large surveys have found frequencies of 18% to 69% for right ventricular hypertrophy with the summation of RV1 and SV5 greater than 10.5 mm.14 As with criteria for LVH, this was met more often in dynamic than static athletes.
Voltage increases of 25% have been noted after only 11 weeks of training, and decreases in voltage are observed following deconditioning, although it occurs more slowly.14 The QRS axis progressively becomes more vertical as athletic conditioning improves, and an incomplete right bundle branch block pattern is not uncommon. This finding has been attributed to an increase in muscle mass at the apex of the right ventricle and has been noted to resolve after cessation of training. Miscellaneous findings include increased P-wave amplitude and notching, as well as prominent U waves. A representative ECG of an individual with changes characteristic of the athletic heart is shown in figure 2.
Clinical Patient Profile
The well-trained athlete exhibits decreased body fat, increased muscle mass, and various cardiovascular changes. The pulse is slowed, and its amplitude may be higher from increased stroke volume. LV impulse duration may be prolonged, and its area enlarged. Grade 1 to 2 systolic murmurs have been noted in 30% to 50% of athletes enagaged in dynamic activities.14 These murmurs have the clinical characteristics of benign flow murmurs. Similarly, third and fourth heart sounds are common. The frequency of third heart sounds is generally greater than 50% in dynamic athletes, with a range of 30% to 100%; fourth heart sounds were somewhat less frequent, varying from 20% to 60%.14
Athletic heart vs hypertrophic cardiomyopathy. Distinguishing between these two conditions is critical because hypertrophic cardiomyopathy (HCM) accounts for a significant number of sudden deaths in athletes during physical activity. Maron et al27 found that 46% of sudden deaths in young athletes were due to HCM or possible HCM (see "Hypertrophic Cardiomyopathy: Practical Steps for Preventing Sudden Death"). Unlike athletic heart syndrome, which reflects the heart's normal adaptation to strenuous physical activity, HCM is characterized by profound hypertrophy, which usually occurs at the expense of the LV cavity, asymmetrical septal hypertrophy, and myofibrillar disarray. Mutations in at least eight genes can cause HCM. Mutations in the beta-cardiac myosin heavy chain occur in only half of HCM patients, suggesting other mutations or nongenetic causes are also responsible.28,29
Noninvasive assessment. Problems frequently arise in the noninvasive evaluation of athletes when dimensions of the LV free wall and septum exceed normal limits. A study30 of childhood swimmers revealed a thickness above the 95th percentile in 81%. Similarly, septal hypertrophy is present in endurance athletes and increases as training progresses.14 Measurements greater than 11 mm were observed in up to 60% of basketball players and 83% of childhood swimmers.30,31 Because of such changes, the ratio of the intraventricular septum to the LV free wall can occasionally exceed the normal ratio of 1.3; it can be as high as 2.0 in athletes.14 Menapace et al32 demonstrated a method to distinguish the physiologic septal hypertrophy that occurs in weight lifters from the pathologic changes in individuals with HCM. Dividing the width of the intraventricular septum by the LV end-systolic diameter clearly distinguishes athletes from those with HCM. A cut-off value of 0.48 or more (the mean plus 3 standard deviations) indicates HCM.
Changes with detraining. Significant overlap can exist in septal or free-wall thickness between athletes and HCM patients. In these cases, discontinuation of training may help distinguish between the athletic heart and HCM. Detraining produces a rapid and progressive decrement in morphology, and repeat echocardiography will show normalized wall thickness in athletes, but not in individuals with HCM.
Mitigating Patients' Concerns
The athletic heart syndrome represents a constellation of clinical findings that arise from normal physiologic adaptation to strenuous physical activity. Several characteristic ECG changes occur that are all the result of increased vagal tone and diminished sympathetic drive; however, distinguishing athletes with athletic heart syndrome from those with other pathologic conditions may require closer scrutiny.
It is critical to communicate clearly with the athlete and, for minors, the parents. Few things can be more alarming than concern about underlying cardiac disease. Thus, when evaluation reveals no evidence of significant disease, the clinician must place this in proper context. Reassurance that such findings are a normal response to training will allow the athlete to return to full activity without worrying that "something may be wrong with my heart."
This article was adapted from the recently published book: Thompson PD (ed): Exercise and Sports Cardiology, New York City, McGraw-Hill Medical Publishing, 2001 (to order: 1-800-262-4729 [ISBN:0-07-134773-9]).
Dr Puffer is a professor and chief of the division of sports medicine in the department of family medicine at the University of California, Los Angeles. Address correspondence to James C. Puffer, MD, 924 Westwood Blvd, Suite 650, Los Angeles, CA 90095-7087; e-mail correspondence to [email protected].
Disclosure information: Dr Puffer discloses no significant relationship with any manufacturer of any commercial product mentioned in this article. No drug is mentioned in this article for an unlabeled use.