Journal Logo

CLINICAL SCIENCES: Clinically Relevant

Electrocardiograms in Athletes

Interpretation and Diagnostic Accuracy

LAWLESS, CHRISTINE E.1; BEST, THOMAS M.2

Author Information
Medicine & Science in Sports & Exercise: May 2008 - Volume 40 - Issue 5 - p 787-798
doi: 10.1249/MSS.0b013e318164dd18
  • Free

Abstract

Attempts to reduce or eliminate sudden cardiac death (SCD) from occurring in young athletes with previously undetected heart disease has led to intense screening efforts to detect the underlying cardiac conditions, with the primary screening tool being the preparticipation history and physical examination (PPE). However, since the routine PPE lacks the sensitivity to reliably detect the main causes of SCD in athletes (22), it has been suggested that cardiac screening tests such as electrocardiography (ECG) and/or echocardiography be added to the standard PPE in order to enhance its ability to diagnose these underlying conditions (7). This appears to be an effective strategy, based on the epidemiologic data obtained from countries where such screening programs have been in place long enough to determine its effects on the incidence of SCD. In Italy, for instance, long standing mandatory cardiac screening of elite athletes appears to have decreased the incidence of SCD by 89% during a 26-yr period (8), while the epidemiology of SCD in this population has shifted towards conditions that are more difficult to diagnose by standard cardiac testing. However, it is not possible from the Italian data to determine how much the ECG alone contributed to screening outcomes, since there is no control group that did not receive ECG screening over the same time period.

Based on the strength of the Italian data, several expert medical consensus groups have published position stands strongly urging that the 12 lead ECG be added to the routine history and physical examination. In 2004, the International Olympic Committee recommended that an ECG be performed on elite athletes prior to Olympic sports participation (6). In 2005, the European Society of Cardiology recommended implementation of a common European screening protocol based on the 12-lead ECG (7). In parallel, participants in the jointly sponsored American Heart Association (AHA)/American College of Cardiology 36th Bethesda Guidelines for Sports Participation concluded that "ECGs are a practical and a cost effective strategic alternative to routine echocardiography for population based preparticipation screening." They also suggested that the ECG will reliably diagnose up to 75-95% of athletes with hypertrophic cardiomyopathy (HCM) (25).

Even though performance of ECG during the PPE is not easily applied to the population of competitive athletes in the United States, where the number may exceed 20 million, there are several initiatives underway to encourage and mandate this strategy. In a survey conducted in 2005 among 122 North American professional sports teams including Major League Baseball, National Hockey League, and National Football League, 92% of the teams reported use of the ECG for preparticipation cardiovascular screening, whereas only 17% of the teams performed exercise testing, and 13% performed echocardiography (15). Since cardiovascular screening and evaluation of the ECG in athletes is becoming more commonplace, it is essential that sports medicine physicians be able to understand the role of ECG screening in their community sports programs and interpret ECG tracings reliably and accurately in the athletic population.

The purpose of this paper is to review the current evidence for the use of ECG in the preparticipation screening of young athletes, as well as ECG interpretation strategies to determine risk of SCD in this population. Two general areas will be discussed: 1) published studies of screening ECG in athletes, especially studies where the ECG was correlated to underlying cardiac structure and/or outcomes; and applicability of these results to specific populations of athletes; and 2) the sensitivity and specificity of the ECG in the various conditions that are known to place athletes at risk for SCD (21): HCM, arrhythmogenic right ventricular dysplasia (ARVD), dilated cardiomyopathy (DCM), myocarditis, long QT syndrome (LQTS) and other channelopathies, anomalous coronary artery, myocardial bridging, aortic stenosis, mitral valve prolapse, and Marfan syndrome or dilated aortic root.

PUBLISHED STUDIES ON ECG IN ATHLETES, AND PROPOSED INTERPRETATION ALGORITHM

There are several excellent references on the use of the ECG for screening large populations of athletes for potential heart disease. The most widely quoted article is by Italian researchers Pelliccia et al. (32), illustrating the clinical significance of various abnormal ECG patterns. Other useful references include those by Fuller, Sharma, and Tanaka (13,38,42) reporting screening ECG in high school student athletes in Nevada, junior elite athletes in the United Kingdom, and Japanese school children, respectively.

Pelliccia and colleagues correlated ECG findings in athletes with underlying cardiac morphology. Between the years 1993 and 1995, 1050 consecutive elite Italian athletes underwent ECG and echocardiogram evaluation. In 45 athletes (4.5%), the echocardiogram or the electrocardiogram was technically suboptimal, leaving a final study cohort of 1005 elite athletes. The mean age was 24 yr, 75% were males, and 38 sports were represented. Only 16 athletes were less than 12 yr of age or greater than 40 yr of age. Length of training was 7 yr; 44% of the athletes competed internationally, and 56% competed nationally. Because all but two athletes were Caucasian, the findings may not necessarily be extrapolated to other ethnic groups such as African Americans.

Pelliccia and colleagues divided the ECG of the 1005 elite athletes into three categories: normal, mildly abnormal, and distinctly abnormal (Table 1). The ECG was considered normal if the R or S wave was ≤ 29 mm, Q waves were absent, T waves were normal, left bundle branch block was absent, the axis was normal, Wolff-Parkinson-White (WPW) pattern was absent, and atrial size was normal. It was possible to have incomplete right bundle branch block present on a normal ECG. Although 60% (603/1005) of the ECGs were considered "normal," there was still a small chance (4%) of having underlying heart disease. Specific cardiac abnormalities that can be associated with normal ECG are discussed in detail below.

TABLE 1
TABLE 1:
Classification scheme for electrocardiograms in athletes (data from Pelliccia et al. (32)).

The ECG was considered "mildly abnormal" if the R or S wave measured 30-34 mm; Q waves up to 2-3 mm in two leads were noted; the T waves were either flat, tall, or mildly inverted in at least two leads; right bundle branch block was present; atrial enlargement was present; the axis was normal; and WPW was absent. Given this ECG pattern, there was a 5% chance of having an underlying cardiac condition. Twenty-six percent of athletes (257/1005) demonstrated this type of ECG.

An ECG was considered "distinctly abnormal" if the R or S waves measured ≥ 35 mm; Q waves were greater than 4 mm in two leads; T waves were deeply inverted greater than 2 mm in two leads; left bundle branch block was present; axis was ≤ −30° or ≥ 110°; or WPW pattern was present. Fourteen percent of athletes (145/1005) demonstrated this type of ECG pattern, and 10% of the 145 were shown to have an underlying cardiac condition. The high rate of underlying cardiac disease in this group may be attributable to the population studied. Pelliccia's cohort consisted of a mixture of 2 types of elite athletes: a group of 785 consecutive elite athletes who underwent routine medical and cardiac screening, and a group of 220 elite athletes specifically referred to the center for suspected cardiac disease. Both the highest incidence of heart disease (15%) and highest incidence of ECG abnormalities (38%) were seen in the latter group of athletes.

The combined power of the mildly abnormal and distinctly abnormal ECG that was noted in about 40% of athletes showed a sensitivity of 51%, specificity of 61%, positive predictive accuracy of 7%, and a very high negative predictive accuracy of 96%. The types of cardiac disease identified in these athletes were mitral valve prolapse, aortic valve diseases, atrial and ventricular septal defects, DCM, pulmonary artery stenosis, myocarditis, HCM, pericarditis, coronary artery disease, WPW syndrome, and hypertension. It is not known how many of these cardiac conditions could have been detected by history and physical examination alone. However, if these data had been analyzed, the added value of the ECG would most likely be less.

All three types of ECG patterns described above varied according to gender, and sport; the abnormal ECG patterns were more likely to be seen in males compared with females, and were more likely to be present in endurance sports such as cycling, cross-country skiing, tennis, canoeing, and basketball (Fig. 1 (32)). In the 145 athletes with distinctly abnormal ECG, the left ventricular end-diastolic volume tended to be higher, as was left ventricular mass index, wall thickness, and left atrial size. Thus, based on this study, there appears to be a correlation between the surface ECG and underlying cardiac adaptation to exercise. This may become important in ECG interpretation in the athlete. For example, if a "distinctly abnormal" pattern is seen in an alpine skier or shooter (where this pattern is very rare), the physician might be more suspicious of an actual underlying cardiac diagnosis, rather than simply attributing this to athletic adaptation. On the other hand, if such a pattern is seen in a basketball athlete (where this pattern is common), the physician might be tempted to attribute it to athletic adaptation and might easily miss the true pathology. More often, there will be a tendency to "overread" the ECG, based on the presence of left ventricular hypertrophy (LVH) alone.

FIGURE 1
FIGURE 1:
Distribution of the three types of ECG patterns according to type of sport among 1005 athletes. ECG that were distinctly abnormal (black bars), mildly abnormal (dark gray bars), and normal or with minor alterations (light gray bars) are depicted as proportions of all the athletes participating in each sporting discipline. Only sports with ≥ 12 participants are shown. X-C, cross-country. Reproduced with permission from Pelliccia A, Maron BJ, Culasso F, et al. Clinical significance of abnormal electrocardiographic patterns in trained athletes. Circulation 2000;102(3):278-84 (32).

Pelliccia's work makes a substantial contribution to our understanding of ECG in athletes. It is apparent that the ECG appearance is dependent on type of sports, and gender. The incidence of underlying heart disease, and abnormal ECG were higher in the referral population, suggesting that it might be important to concentrate screening efforts on the symptomatic athlete, or the one with abnormal history and physical. However, this work has its limitations. The cohort was heterogenous, and it contained a mixture of two cohorts, including a referred population. Thus, this data cannot be compared directly with other large ECG screening studies. Pelliccia provides detailed sensitivity and specificity for the ECG as a screening tool, but from his data we cannot determine what value the ECG added beyond the history and physical alone. Finally, the data cannot be directly applied to those of African origin, such as Afro Caribbeans or African Americans, since large numbers of these types of athletes were not included in this study.

It is important to remember that the Italian data were derived from a population of elite white European athletes with a mean age of 24 yr. Since ECG patterns are quite different in the school age child and adolescent (11), and phenotypic expression of the genetic diseases that are known to cause SCD in young athletes may be absent until the teenage years (23), the Italian findings and recommendations may not necessarily be directly applied to athletes below age 17 yr, or to different ethnic groups. It would therefore seem imperative that we understand normal and abnormal ECG patterns in athletes of all ages, genders and ethnic groups, particularly if earlier intervention to detect those at risk for SCD is to be successful.

In 1999, Sharma and colleagues were the first to systematically address the issue of the younger athlete, by studying a group of 1000 junior elite British athletes with a mean age of approximately 15 yr (38). The majority of these athletes (988) were white, 8 were Afro Caribbean, and 8 were Asian. The sports represented included tennis, football (soccer), rugby, cycling, swimming, athletics, boxing, rowing, and triathlon. A variety of ECG abnormalities were noted (and absent or seen at a lesser frequency in 300 control, nonathletic, age-matched individuals). Findings more often detected in junior elite athletes included sinus bradycardia (80%), sinus arrhythmia (52%), first-degree AV block (5%), incomplete right bundle branch blocks (29%), left atrial enlargement (14%), right atrial enlargement (16%), ST segment elevation (45%), tall peaked T waves (22%), and isolated Sokolow voltage criteria for LVH (45%). Corrected Q-T interval and QRS duration were longer in the athletes compared with nonathletes. None of the athletes in this study were found to have significant underlying pathology. However, based on the absence of the following features in the study population of young athletes, and the known association of these features with cardiac pathology, the authors concluded that presence of such ECG abnormalities may be indicative of underlying pathology in a highly trained junior athlete: ST depression or deep T inversion; minor T wave inversions in any lead except V2-V3 when athlete is less than 16 yr old; Romhilt-Estes voltage criteria for LVH in female athletes; pathological Q wave; left-axis deviation; or complete left bundle branch block.

Sharma's work contributed to what we know about ECG in the young athlete by defining differences in the adolescent athletes ECG compared to a control group of nonathletes. Although the findings in junior athletes appear to be similar to what is seen in adults, we do not have a direct comparison between these groups. Similar to Pelliccia's work, the population is also primarily Caucasian, and the findings cannot necessarily be extrapolated to other ethnic groups.

In 1997, Fuller prospectively screened 5615 high school athletes in northern Nevada for risk of SCD (14). The high school student athletes were screened with a PPE and an ECG and were followed up with cardiac testing if there were any positive findings, defined as abnormal history or physical examination and/or abnormal ECG (bundle branch block, premature beats, tachycardias, preexcitation, high-degree AV block, prolonged QT interval, LVH, Q waves, ST-T abnormalities). For the purposes of this study, cardiac outcome measures were defined as detection of any cardiovascular disease during the screening process that would preclude sports participation according to the 16th Bethesda guidelines. It is important to note that because the cohort was not followed for incidence of adverse cardiac events, adverse events were not included in the definition of a cardiac outcome. Results showed that 5% were found to have R or S waves greater than 30 mm; 6% had T wave flattening or inversion in two or more leads; 2% had abnormal Q waves; and the axis was deviated greater than −30° or 120° in 1%. ECG abnormalities were present in 15.7% of the entire cohort. Cardiac history led to detection of outcome measures in 0 athletes, auscultation/inspection in 1/6000 athletes, blood pressure measurement in 1/1000 athletes, and the ECG in 1/350 athletes. In total, cardiac outcome measures were determined in 22 athletes, or one out of every 255 athletes. Thus, the sensitivity of the PPE to detect the abnormality was only 6%. However, the use of ECG significantly increased the sensitivity to 70%, while echocardiography increased it further to 80%. It is important to note that echocardiography was not performed in all subjects, and 15/22 athletes with outcome measures were lost to follow-up. In another article, the authors approximated the cost to perform each test and to evaluate abnormal screening findings. Years of life gained through detection of athletes with potential causes of sudden cardiac death were estimated. Overall, the cost per estimated year of life saved with the use of PPE plus ECG was calculated to be $44,000 compared with $84,000 for the PPE alone, and $200,000 if an echocardiogram had been performed in all cases (13).

Fuller's study differed from both Pelliccia's and Sharma's in that the cohort was larger and comprised American high school athletes, rather than European elite junior or senior athletes. Because echocardiography was not performed in the entire cohort, the true incidence of underlying cardiac disease in Fuller's cohort is not known. However, through use of the combined PPE and ECG, with subsequent evaluation of abnormals, the incidence of underlying heart disease appeared to be about 0.3%. The cohort was virtually all Caucasian, so the data cannot necessarily be applied to other ethnic groups, such as African Americans. The three ECG studies cited above cannot be compared directly because of differences in population studied, and differences in methodology. However, one can conclude that the incidence of lethal underlying heart disease in the young athletic population is actually quite low, probably in the range of 0.3% or less. The higher incidence seen in Pelliccia's work (4% to as high as 10% in those with distinctly abnormal ECG) was most likely due to the fact that he used a different definition of a "cardiac outcome," and he included the referred population in the analysis.

All three of the above studies have focused on use of the ECG to either correlate surface ECG findings with underlying cardiac structure, or to correlate the screening ECG with subsequent cardiac diagnoses, as determined by further cardiac investigations. None of the studies were primarily designed to prospectively evaluate the ECG in its ability to add value to a standardized PPE, nor to determine the effect of the screening ECG on overall cardiac outcomes, survival, cost, procedural complications, and impact on the athlete. In Europe, the recommendation to advise routine ECG screening as part of the PPE was based primarily on data generated by Italian researchers based in the Veneto region of Italy (8). In 1982, Italy passed a nationwide law requiring that all competitive athletes between the ages of 12 and 35 yr undergo complete evaluation prior to sports participation. The evaluation included both a standardized PPE and a 12-lead ECG. Sudden cardiac death rates for the years 1979-1981 (prescreening), 1982-1992 (early screening), and 1993-2004 (late screening) were analyzed, and the rates in screened athletes were compared with those seen in unscreened nonathletes during the same eras in the Veneto region. The results demonstrated a striking 89% decrease in the annual incidence of SCD in athletes, due mainly to a reduction in the SCD rate from cardiomyopathies. This was accompanied by a similar trend in the disqualification rate from cardiomyopathies seen in a subset of screened athletes (11%) at the Padua Sports Medicine Center. The study had its limitations. This was a nonrandomized, retrospective study, and although the ECG was felt to be responsible for the trends noted, it is possible that the favorable trends may have been due to other factors such as meticulous application of the PPE by highly trained "screening physicians." The comparator group of athletes who survived but were disqualified from sports participation represented only 11% of all the athletes in Veneto, and they were seen only at one sports medicine center in one city, Padua (8). Thus, the comparison between this subset and the overall group may not be valid.

In 1973, a national screening system for cardiovascular diseases was introduced in Japan for all 1st-, 7th-, and 10th-grade students, using both a questionnaire and a 12-lead ECG (42). In this cohort of both athletes and nonathletes with previously undiagnosed disease, 1876/37,807 students were found to have an abnormality on preliminary questionnaire/ECG screening. Of these, 9/1876 students, or 0.024%, were determined to have underlying high-risk conditions: five HCM, one left ventricular dilatation, one primary pulmonary hypertension, one WPW syndrome, and one long-QT syndrome. Of the remaining students, 497/1876 were found to have low-risk cardiac conditions, while the remaining 1370/1876 were diagnosed as normal on further testing. The authors concluded that their screening strategy was highly effective for finding disease, but that not all deaths could be prevented. There were three SCD during the follow-up period, all occurring during exercise. One occurred in a child with HCM detected through screening (classified as "high risk") who continued to exercise against medical advice. In the other two, initial questionnaires/ECG were normal (classified as "normal"). Postmortem reviews of their ECG showed slight right-axis deviation. But, autopsies were not obtained, so exact causes of SCD were not determined.

Aside from analyzing the ECG for the presence of LVH, LQTS, or ventricular preexcitation, an athlete may also experience either premature ventricular contractions (PVCs) or complex ventricular ectopy on a resting ECG. Athletes with either of these conditions may or may not be symptomatic, at rest or with exercise. Several studies have addressed the likelihood of underlying heart disease given the presence of ventricular ectopy on resting ECG. Long-distance runners, in particular, may frequently demonstrate this finding (41). Of 15,889 highly trained Italian athletes screened for cardiac conditions (5), up to 2.2% experienced palpitations due to a ventricular rhythm disturbance or demonstrated greater than or equal to three PVCs on a resting ECG. The chance of underlying cardiac disease given these findings on ECG is 7%. However, athletes can be further differentiated for the risk of underlying heart disease through the use of 24-h ambulatory monitoring to quantify the PVC burden, with the risk increasing with higher PVC burden. The presence of more than 2000 PVCs during 24 h indicates a 30% risk of underlying heart disease, whereas having fewer than 2000 PVCs during 24 h markedly reduces this risk to less than 5% (5).

Despite the compelling Italian data, endorsements by major worldwide professional societies, and the adoption of the preparticipation screening ECG by many athletic organizations, ECG screening of athletes has its critics, and its limitations (24,44). Although the nonrandomized Italian observational study demonstrated a substantial decline in SCD rate over the study period (8), as noted above, it is entirely possible that the observations noted were due to other factors, such as meticulous application of the PPE by specialized sports physicians, or geographic differences in the two groups of athletes being compared (8,44).

Accurate, rigorous evaluation of the effects of the ECG on the PPE screening process would require a prospective randomized study, comparing standardized PPE, including the 12 AHA questions, with the standardized PPE plus the 12-lead ECG. There would have to be some "gold standard" as to whether or not underlying heart disease is present. This "gold standard" would be a test capable of detecting subtle forms of cardiomyopathy, as well as coronary anomalies. The echocardiogram is practical, but it may not be reliable in detecting unusual forms of hypertrophic and right ventricular cardiomyopathy, and/or coronary arterial anomalies. Outcomes would have to include a variety of primary and secondary outcomes, including overall incidence of significant underlying heart disease; differences in ECG between different ethnic groups, ages, and genders; cost-benefit analysis of both the initial ECG and subsequent cardiac investigation; impact of lost training time; and impact of procedural-rated complications. For instance, if an athlete with HCM requires an implanted defibrillator, what is the impact of an infected lead on overall outcomes?

To summarize, the ECG appears to be an inexpensive, reasonable screening modality to detect underlying cardiac disease. However, it must be made clear that detecting underlying cardiac disease may not necessarily translate into saving lives, or preventing SCD. The reason for this is that detection of disease, or of probable disease may lead to institution of costly, life-saving therapies such as the implantable defibrillator. These treatments come with significant complication rates; thus, any comprehensive cardiovascular screening program or study must include an analysis of any negative outcomes that might result from such complications. It is entirely possible that such negative outcomes might offset any potential reduction in SCD due to the underlying disease, and may actually increase both the cost of care and overall mortality.

Nonetheless, if ECG screening is chosen, it appears to have several distinct advantages over other screening techniques. It significantly improves the sensitivity of the PPE screening examination; it has reasonable specificity and excellent negative predictive value; it is inexpensive compared to the echocardiogram; and many primary care and sports medicine physicians, as well as cardiologists are already quite comfortable with its interpretation in the general population. Although there is no uniform algorithm for interpretation of the ECG in asymptomatic athletes, the Italian screening program has proposed some simple rules that will reliably detect the majority of heart disease in young athletic populations (Table 1 (32)). It is important to note that the Italian ECG patterns are based on results obtained in elite Caucasian European athletes with a mean age of 24 yr, and that rules of ECG interpretation derived from the Italian data may not necessarily be applied to younger athletes, to the recreational athlete or to athletes of other ethnic groups. One such group is likely African American athletes. Based on published data in nonathletic African American populations, there is reason to suspect that ECG in athletic African Americans may be markedly different than in their white European counterparts (20,32,33,43). Further study is necessary to determine normative data in this group of athletes.

As there is currently no standard for ECG interpretation in the athlete, the ECG is subject to marked variation in interpretation by cardiologists, and others involved in athlete care. It is not 100% sensitive for the detection of underlying heart disease; hence, the possibility of false-negatives exists. False-positives exist as well, as normal athletic adaptation to exercise can result in marked alterations of the surface ECG, resulting in delays due to the time it takes to follow up and perform further cardiac evaluation. And even though it appears to be a practical, cost-effective alternative in most clinical settings, and the European cardiology community has widely endorsed ECG screening (7), a recent AHA expert consensus panel did not recommend its widespread adoption as a screening method for athletes in the United States (24). The rationale for this decision included: the cost of screening such a large number of eligible athletes; the high false-positive rate and the cost of evaluating such false-positives; lack of trained "screening physicians" in the United States as compared with Italy; lack of a randomized trial demonstrating clear superiority of the ECG over the PPE; lack of a clear standard for interpretation of ECG in athletes; lack of normative data in certain demographic and ethnic groups; the likelihood that asymptomatic athletes with underlying lethal conditions might be very different from symptomatic individuals with the same conditions; and the possibility that ECG screening might actually increase the death rate, via evaluation and treatment-related procedural complications.

Despite the critiques and limitations of the ECG, for those clinicians who choose to use an ECG-based approach in their preparticipation screening of athletes, the decision tree shown in Figure 2 may be a useful guide. Our proposed approach is consistent with what the Europeans have recommended (7). The athlete with the distinctly abnormal ECG requires further investigation, regardless of symptoms. Cardiac work-up is considered "optional" for the athlete without symptoms and mildly abnormal ECG, or the one with positive answers to the AHA questions, but totally normal ECG.

FIGURE 2
FIGURE 2:
Proposed decision tree for ECG-based cardiovascular screening of athletes.

RELIABILITY OF THE ECG IN THE VARIOUS CONDITIONS THAT ARE KNOWN TO CAUSE SCD IN ATHLETES

The studies quoted above indicate that the ECG is about 50-95% sensitive in its ability to detect underlying heart disease (25,32). While this represents a marked improvement over physical examination alone, where the sensitivity is about 3-6% (14,22), the ECG cannot be expected to reliably detect all forms of heart disease that might lead to SCD in athletes, because of its variable sensitivity for these conditions (Table 2).

TABLE 2
TABLE 2:
Sensitivity and specificity of electrocardiogram (ECG) for specific cardiovascular diagnoses.

ECG in Myocardial Diseases

Classic ECG features of HCM include LVH, repolarization abnormalities, left-axis deviation, and abnormal Q waves (Fig. 3), with all features being more prevalent in the obstructive form of the disease (34). However, since the chance of an abnormal ECG in a patient who is asymptomatic without obstruction (73%) is somewhat less than in a patient with symptomatic obstruction (98%) (34), there may be up to a 25% chance of a normal ECG (false-negative) in an asymptomatic athlete without a cardiac murmur. ARVD causes anatomic damage with corresponding alterations of electrical activation and depolarization (27,29). Although a rare cause of SCD in athletes in the United States (21), characteristic ECG features of this condition with their corresponding prevalence include T wave inversions in V1-V3 (55-94%), QRS duration ≥ 110 ms in V1-V3 (64%), and the presence of an epsilon wave (electric potentials after the end of the QRS complex; 25-33%) (Fig. 4) (27,29). Since inverted T waves are commonly seen in children less than 12 yr of age and in normal individuals, this feature is not specific enough to make a diagnosis. The ECG is always virtually always abnormal in DCM. Features are nonspecific, but include LVH, low ORS voltage, left bundle branch block in at least 20% of cases, premature beats or nonsinus rhythm, and ST-T abnormalities (39). Myocarditis accounts for 5% of cases of SCD in athletes (21), up to 20% of SCD in young military recruits (12), and may be seen in up to 10% of those suffering from acute influenza infection (19,30). ECG findings of acute myopericarditis can mimic acute myocardial ischemia (9) (Fig. 5) and may include ST segment elevation in two or more leads (54%), T wave inversions (27%), widespread ST segment depressions (18%), and pathological Q waves (18-27%) (9).

FIGURE 3
FIGURE 3:
Typical ECG seen in hypertrophic cardiomyopathy (HCM). Note the left-axis deviation, abnormal Q waves, and left ventricular hypertrophy (LVH) in panel A, and the premature ventricular beat, marked ST-T abnormalities, and LVH in panel B. (ECGs courtesy of Dr. Christine Lawless.)
FIGURE 4
FIGURE 4:
ECG of arrhythmogenic right ventricular dysplasia (ARVD). Note the characteristic epsilon waves (electric potentials in the ST segment, black arrow) in V2. (ECG courtesy of Dr. Raul Weiss.)
FIGURE 5
FIGURE 5:
A 24-yr-old cyclist presented with a 10-d history of progressive chest pain. ECG showed 3- to 4-mm ST elevation in two, three, AVF, and lateral leads, suggesting either acute ischemia or focal myopericarditis. Immediate cardiac catheterization showed inferior hypokinesia and normal coronaries. Cardiac biopsy confirmed lymphocytic infiltrate and myocyte necrosis consistent with acute myocarditis. (ECG courtesy of Dr. Ed Michl, and Dr. Steve Lieberman.)

ECG in Ion Channelopathies

LQTS accounts for only 1-2% of the SCD in athletes (21), but it is easily detected by ECG. The QT interval is typically measured in lead II of a 12-lead ECG and corrected for heart rate by using Bazett's formula (QTc = QT/√RR). A QTc interval of >440 ms in men or > 460 ms in women is considered prolonged (Fig. 6). However, Bazett's formula has been criticized as inaccurate, especially at heart rate extremes, and may overestimate the QT interval during bradycardias, which frequently occur in athletes due to high vagal tone at rest (18). Each of the genetic types of LQTS has one specific trigger for cardiac events, with LQTS1 being the one most often triggered by exertion (35,36). Although 83-100% of gene carriers manifest QTc of ≥ 440 ms on ECG, in those with suspected LQTS1 who do not demonstrate a prolonged QTc, exercise testing or alternatively, epinephrine infusion may prove helpful in unmasking the prolonged QTc (1,40). Brugada syndrome, first described in 1992, is characterized on ECG by the presence of right bundle branch block, and "coved" ST elevation in leads V1-V3 (Fig. 7) (3,26,46). Brugada can be difficult to distinguish from ARVD or other causes of right bundle branch block and ST elevation.

FIGURE 6
FIGURE 6:
ECG in LQTS1, and Bazett's formula for calculation of corrected Q-T interval. (ECG courtesy of Dr. Raul Weiss.)
FIGURE 7
FIGURE 7:
Brugada syndrome ECG showing typical "coved" appearance of the ST segment and the complete RBBB. (ECG courtesy of Dr. Raul Weiss.)

ECG in Congenital Forms of Coronary Artery Disease, Valvular Disease, and Aortic Disease

In general, anomalous coronary artery disease has no characteristic ECG features, and, in most instances, even in those who have died during athletics, the ECG has been completely normal (2,4). Athletes with myocardial bridging may present with exercise-induced chest pain or ischemia, conduction abnormalities, rhythm disturbances, left ventricular dysfunction, acute myocardial infarction, or sudden cardiac death (28), but, in most instances, the resting ECG is normal (28). The characteristic ECG features of aortic stenosis are LVH and ST-T abnormalities, which can be seen in up to 80% of younger individuals and probably even higher in older individuals (37,45). Mitral valve prolapse (MVP) can be difficult to understand clinically, as there appear to be two types (17). Anatomic MVP is classified as an inheritable connective tissue disorder, characterized by myxomatous degeneration of the mitral valve, with an increase in the proteoglycan material, breakdown of normal collagen, and production of abnormal collagen fibrils. Echocardiography shows a thickened, redundant mitral valve, with varying degrees of mitral regurgitation noted (17). The anatomic form can also lead to serious arrhythmias and SCD (31), cerebral embolism, and infective endocarditis. There have been no large series of the type of MVP associated with SCD in athletes, but one might assume from the available literature that those with the anatomic form of the disease would be considered more at risk. In the syndromic form of MVP, the valve itself appears to be normal, and is not myxomatous. Echocardiography confirms this, but may or may not show mild "bowing" of the valve, without thickening or redundancy. ECG findings vary in MVP, and are nonspecific. Most common findings are T wave inversions or ST segment depressions, which are seen in about one third of MVP cases, and premature ventricular contractions in another one third (10,17). Rarely, there may be associated QT prolongation (10,17). Considering all forms of MVP, a normal ECG is likely the rule, rather than the exception. Occasionally, MVP can be associated with diseases of the aorta such as Marfan syndrome or other fibrillinopathies. Although not always fatal (16), aortic dissections or ruptures due to these conditions account for 3% of SCD in athletes. ECG is unlikely to be abnormal in most cases of Marfan syndrome or related conditions. The diagnosis is still primarily a clinical diagnosis.

SUMMARY

In summary, the ECG is evolving as an adjunct to the standard PPE in screening young athletes for cardiac conditions. Although the literature is lacking well-controlled prospective trials, the sensitivity in detecting underlying cardiac disease in this population is about 51-95%, depending on the type of underlying disease and the population being studied. Since many of the conditions that cause SCD in athletes demonstrate similar ECG findings as to what is seen in normal athletic adaptation, clinicians need to follow some simple rules in ECG interpretation in athletes, and they need to be prepared for the consequences of both over- and underinterpretation of the ECG in this group. Although ECG-based screening has gained wide acceptance in Europe, ECG-based cardiovascular screening of young athletes is currently not recommended in the United States, for numerous reasons. Forthcoming studies on high-risk populations for SCD, such as young, African American athletes, are needed to determine normative baseline data and adaptive changes to exercise.

The authors are grateful for the review of this manuscript by Dr. Antonio Pelliccia of the Institute of Sports Science in Rome, Italy.

REFERENCES

1. Ackerman MJ, Khositseth A, Tester DJ, Hejlik J, Shen WK, Porter CJ. Epinephrine-induced QT interval prolongation: a gene-pecific paradoxical response in congenital long QT syndrome. Mayo Clin Proc. 2002;77:413-21.
2. Angelini P, Velasco JA, Flamm S. Coronary anomalies: incidence, pathophysiology, and clinical relevance. Circulation. 2002;105:2449-54.
3. Antzelevitch C. Brugada syndrome. Pacing Clin Electrophysiol. 2006;29(10):1130-59.
4. Basso C, Maron BJ, Corrado D, Thiene G. Clinical profile of congenital coronary artery anomalies with origin from the wrong aortic sinus leading to sudden death in young competitive athletes. J Am Coll Cardiol. 2000;35:1493-501.
5. Biffi A, Pelliccia A, Verdile L, et al. Long-term clinical significance of frequent and complex ventricular tachyarrhythmias in trained athletes. J Am Coll Cardiol. 2002;40:446-52.
6. Bille KA, Figueiras DB, Schamasch PC, et al. A sudden cardiac death in athletes: the Lausanne recommendations. Eur J Cardiovasc Prev Rehabil. 2006;13(6):859-75.
7. Corrado D, Pelliccia A, Bjornstad H, et al. Cardiovascular pre-participation screening of young competitive athletes for prevention of sudden death: proposal for a common European protocol. Consensus statement of the study group of sport cardiology of the working group of cardiac rehabilitation and exercise physiology and the working group of myocardial and pericardial diseases of the European Society of Cardiology. Eur Heart J. 2005;26(5):516-24.
8. Corrado D, Basso C, Pavei A, Michieli P, Schiavon M, Thiene G. Trends in sudden cardiovascular death in young competitive athletes after implementation of a preparticipation screening program. JAMA. 2006;296:1593-601.
9. Dec GW, Waldman H, Southern J, Fallon JT, Hutter AM, Palacios I. Viral myocarditis mimicking acute myocardial infarction. J Am Coll Cardiol. 1992;20:85-9.
10. Devereux RB, Perloff JK, Reichek N, Josephson ME. Mitral valve prolapse. Circulation. 1976;54:3-14.
11. Dickinson D. The normal ECG in childhood and adolescence. Heart. 2005;91(12):1626-30.
12. Eckart R, Scoville S, Campbell C, et al. Sudden death in young adults: a 25-year review of autopsies in military recruits. Ann Intern Med. 2004;141(11):829-34.
13. Fuller CM. Cost effectiveness analysis of screening of high school athletes for risk of sudden cardiac death. Med Sci Sports Exerc. 2000;32(5):887-90.
14. Fuller CM, McNulty CM, Spring DA, et al. Prospective screening of 5,615 high school athletes for risk of sudden cardiac death. Med Sci Sports Exerc. 1997;29(9):1131-8.
15. Harris KM, Sponsel A, Hutter AM, Maron BJ. Brief communication: cardiovascular screening practices of major North American professional sports teams. Ann Intern Med. 2006;145:507-11.
16. Hatzaras I, Tranquilli M, Coady M, Barrett PM, Bible J, Elefteriades JA. Weight lifting and aortic dissection: more evidence for a connection. Cardiology. 2007;107:103-6.
17. Jacobs W, Chamoun A, Stouffer GA. Mitral valve prolapse: a review of the literature. Am J Med Sci. 2001;320:401-10.
18. Kapetanopoulos A, Kluger J, Maron B, Thompson P. The congenital long QT syndrome and implications for young athletes. Med Sci Sports Exerc. 2006;38(5):816-25.
19. Karjalainen J, Nieminen MS, Heikkila J. Influenza A1 myocarditis in conscripts. Acta Med Scand. 1980;207:27-30.
20. Lewis JF, Maron BJ, Diggs JA, Spencer JE, Mehrotra PP, Curry CL. Preparticipation echocardiographic screening for cardiovascular disease in a large, predominantly black population of collegiate athletes. Am J Cardiol. 1989;64(16):1029-33.
21. Maron BJ. Medical progress: sudden death in young athletes. N Eng J Med. 2003;349:1064-75.
22. Maron BJ, Shirani J, Poliac LC, Mathenge R, Boberts WC, Mueller FO. Sudden death in young competitive athletes. Clinical, demographics, and pathological profiles. JAMA. 1996;276:199-204.
23. Maron BJ, Spirito P, Wesley Y, Arce J. Development and progression of left ventricular hypertrophy in children with hypertrophic cardiomyopathy. N Eng J Med. 1986;315:610-4.
24. Maron BJ, Thompson P, Ackerman M, et al. Recommendations and considerations related to preparticipation screening for cardiovascular abnormalities in competitive athletes: 2007 update: a scientific statement from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism: endorsed by the American College of Cardiology Foundation. Circulation. 2007;115(12):1643-55.
25. Maron BJ, Zipes DP. Introduction: eligibility recommendations for competitive athletes with cardiovascular abnormalities-general considerations. J Am Coll Cardiol. 2005;45:1318-21.
26. Matsuo K, Masazumi A, Nakashima E, et al. The prevalence, incidence and prognostic value of the Brugada-type electrocardiogram. A population-based study of four decades. J Am Coll Cardiol. 2001;38:765-70.
27. McKenna WJ, Thiene G, Nava A, et al. Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task force of the working group on myocardial and pericardial disease of the European Society of Cardiology and of the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. Br Heart J. 1994;71:215-8.
28. Mohlenkamp S, Hort W, Junbo G, Erbel R. Update on myocardial bridging. Circulation. 2002;106:2616-22.
29. Nasir K, Bomma C, Harikrishna T, et al. Electrocardiographic features of arrhythmogenic right ventricular dysplasia/cardiomyopathy according to disease severity. Circulation. 2004;110:1527-34.
30. Nicholson KG, Webster R, Hay A. Human influenza. In: Nicholson KG, Webster R, Hay A, editors. Textbook of Influenza. Oxford (UK): Blackwell Science; 1997. p. 219-63.
31. Nishimura RA, McGoon M, Shub C, Miller FA Jr, Ilstrup DM, Tajik AJ. Echocardiographically documented mitral-valve prolapse: long-term follow-up of 237 patients. N Eng J Med. 1985;313:1305-9.
32. Pelliccia A, Maron BJ, Culasso F, et al. Clinical significance of abnormal electrocardiographic patterns in trained athletes. Circulation. 2000;102(3):278-84.
33. Rao PS. Racial differences in electrocardiograms and vectorcardiograms between black and white adolescents. J Electrocardiol. 1985;18:309-14.
34. Savage D, Seides S, Clark C, et al. Electrocardiographic findings in patients with obstructive and non-obstructive hypertrophic cardiomyopathy. Circulation. 1978;58:402-8.
35. Schwartz PJ, Locati EH, Napolitano C, Priori SG. The long QT syndrome. In: Zipes DP, Jalife J, editors. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia (PA): WB Saunders Co; 1995. p. 788-811.
36. Schwartz PJ, Priori SG, Spazzolini C, et al. Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for life-threatening arrhythmias. Circulation. 2001;103(1):89-95.
37. Seiler C, Jenni R. Severe aortic stenosis without left ventricular hypertrophy: prevalence, predictors, and short-term follow up after aortic valve replacement. Heart. 1996;76(3):250-5.
38. Sharma S, Whyte G, Elliott P, et al. Electrocardiographic changes in 1000 highly trained junior elite athletes. Br J Sports Med. 1999;33:319-24.
39. Shaver, JA, Brest A. Cardiomyopathies: clinical presentation, differential diagnosis, and management. IN: Cardiovascular Clinics. Philadelphia (PA): FA Davis Company; 1988. p 84-117.
40. 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;107:838-44.
41. Talan DA, Bauernfeind RA, Ashley WW, Kanakis C Jr, Rosen KM. Twenty-four hour continuous ECG recordings in long-distance runners. Chest. 1982;82:19-24.
42. Tanaka Y, Yoshinaga M, Anan R, et al. Usefulness and cost effectiveness of cardiovascular screening of young adolescents. Med Sci Sports Exerc. 2006;38(1):2-6.
43. Thapar MK, Harp RJ. Racial variations in electrocardiograms and vectorcardiograms between black and white children and their genesis. J Electrocardiol. 1984;17:239.
44. Thompson PD, Levine B. Protecting athletes from sudden cardiac death. JAMA. 2006;296(13):1648-50.
45. Tveter KJ, Foker JE, Moller JH, Ring WS, Lillehei CW, Varco RL. Long-term evaluation of aortic valvotomy for congenital aortic stenosis. Ann Surg. 1987;206(4):496-503.
46. Viskin S, Fish R, Eldar M. Prevalence of the Brugada sign in idiopathic ventricular fibrillation and healthy controls. Heart. 2000;84:31-6.
Keywords:

PREPARTICIPATION EXAMINATION; CARDIOVASCULAR SCREENING; ATHLETE EVALUATION; CARDIOVASCULAR DISEASE; SUDDEN CARDIAC DEATH IN ATHLETES

©2008The American College of Sports Medicine