Preparticipation Cardiovascular Screening of Young Athletes
An Epidemiologic Perspective
Dennis Y. Wen, MD
THE PHYSICIAN AND SPORTSMEDICINE - VOL 33 - NO. 12 - DECEMBER 2021
In Brief: Sudden cardiac death in young competitive athletes is a rare—but often highly publicized—event. Preparticipation screening for potentially lethal cardiovascular conditions is a complicated and controversial issue. Clinicians need a firm grasp of sound epidemiologic principles concerning screening programs when using the history and physical examination, echocardiography, and electrocardiography to detect potential problems. The very low prevalence of potentially lethal conditions and the low sensitivity and low specificity of the available methods of detection, used alone or in combination, render the positive predictive value of these screening methods exceedingly low.
The sudden death of a young, seemingly healthy, athlete is a rare event that often draws substantial attention. Interest in preparticipation screening for potentially lethal conditions in young athletes has risen in recent years.1 No true consensus exists on this difficult topic, and emotional issues can often override practical medical perspectives. Accepted criteria for an ideal screening program have been developed,2 but the epidemiologic perspective on issues concerning cardiovascular screening of athletes for potentially lethal conditions is often omitted in debates on this topic.
The incidence of sudden cardiac death in young athletes is unknown, although several estimates exist. Ragosta et al3 estimated 1 death per year in 280,000 recreational athletes younger than 30 in Rhode Island. Van Camp et al4 estimated 1 death per 300,000 high school and college athletes, or fewer than 10 deaths per year in the entire nation. Maron et al5 estimated an incidence of 1 sudden cardiac death in 217,400 high school athletes per year in Minnesota.
Causes of Sudden Cardiac DeathTwo large case studies by Van Camp et al4 and Maron et al6 examined deaths in young athletes involved in organized sports in the United States (table 1). Both studies found that hypertrophic cardiomyopathy (HCM) and probable HCM accounted for approximately 50% of the deaths, followed by coronary artery anomalies, coronary artery disease, aortic ruptures, aortic stenosis, myocarditis, dilated cardiomyopathy, arrhythmogenic right ventricular dysplasia (ARVD), and several others. The authors of both studies acknowledged that some causes, such as long QT syndrome or other dysrhythmias, may have been impossible to diagnose postmortem.
Hypertrophic cardiomyopathy. The understanding of HCM has increased greatly in recent years, although it remains incompletely understood. HCM is heterogenous in its causes, clinical manifestations, pathologic findings, and prognosis. At least 150 mutations in 10 different sarcomeric proteins that lead to HCM are now recognized.7,8 The inheritance pattern of HCM is autosomal dominant, but incomplete penetrance exists.
Sudden death in patients who have HCM is presumably caused by dysrhythmias, usually ventricular fibrillation or ventricular tachycardia. The exact prognostic factors for susceptibility to dysrhythmias, and therefore susceptibility to sudden death, are currently unknown. Greater degrees of myocardial hypertrophy confer greater risk of sudden death,7,9,10 but even individuals with minimal hypertrophy can suffer lethal dysrhythmias.7,10-12 Several studies10,13-17 of large family kindreds with HCM have documented differing survival curves for differing mutations, but the clinical utility of genetic analysis for HCM is limited, and the implications for most mutations remain unknown.
Symptoms can vary widely and are not always correlated with the degree of hypertrophy or genetic mutation. HCM can often be completely asymptomatic, with sudden death as the first symptom. Symptoms, when present, can include chest pains, dyspnea, exercise intolerance, syncope, and palpitations. Family history of HCM may be observed, but because of the known incomplete penetrance of HCM and the existence of sporadic cases, family history is not always positive.7,11 Physical examination findings are often few or nonexistent. The classic aortic outflow murmur is usually not present, because only a minority of HCM patients have subaortic stenosis, but this does not necessarily predict a good prognosis.7 Therefore, history and physical examination can often be unreliable as a means for detecting HCM.
The gold standard for diagnosing HCM is echocardiography.7,18-20 However, the lack of complete morphologic manifestation until full maturity can make echocardiography somewhat unreliable. Clearly, a subset of patients who have HCM are at sufficiently low risk of developing lethal dysrhythmias, such that restriction from athletics is unwarranted. Identifying these low-risk athletes is problematic. It is not certain which risk factor, or combination of risk factors, confer a high enough risk to warrant restriction from sports.
Anomalies. Various coronary artery anomalies make up the second most common category of sudden cardiac death in young US athletes.4,6 Several congenital anomalies of the coronary arteries are known, with varying degrees of lethality.21 The anomalies that may lead to sudden death in young athletes usually involve either the left or right coronary arteries originating from the right or left sinuses of Valsalva, respectively. This disorder means the arteries must course between the aorta and the pulmonary trunk. One hypothesis for sudden death is that during vigorous physical exertion, the dynamic increase in volume and pressure of the great vessels mechanically compress the anomalous coronary artery passing between them. Such a phenomenon leads to ischemia and possibly lethal dysrhythmia.21,22 Another hypothesis, considered more likely, is that the sharp angle of take-off at the origin of the anomalous coronary artery from the leaflet produces low flow into the coronary artery under vigorous dynamic conditions.21
Symptoms are highly variable and can include exercise intolerance, exertional chest pains, dyspnea, and syncope. Like HCM, the conditions can be asymptomatic, and sudden death may be the first symptom. Physical exam findings may be nonexistent; therefore, history and physical exam can be an unreliable means of detecting anomalous coronary arteries. Coronary angiography is the most reliable means of establishing the diagnosis, although echocardiography, especially transesophageal, can suggest the diagnosis.
Myocarditis. Another frequently mentioned cause of sudden cardiac death in young athletes is myocarditis.4,6,23,24 Numerous purported causes have been identified, including various viruses, bacteria, parasites, toxins, medications, and drug abuse. The incidence is unknown, and numerous mild or subclinical cases may exist. Sudden cardiac death is thought to occur from lethal dysrhythmias arising from an unstable myocardial substrate in active or healed lesions. Symptoms are variable to nonexistent and can include dyspnea, palpitations, chest pains, and syncope. Echocardiography variably suggests the diagnosis.
Arrhythmogenic right ventricular dysplasia. Although ARVD has been implicated as the most common cause of sudden death in Italian athletes,25 it is considered relatively rare, but perhaps underdiagnosed, in the United States.4,5 Symptoms can vary and may not always occur. The ability of echocardiography or electrocardiography (ECG) to accurately detect ARVD is unclear, and magnetic resonance imaging is considered by some to be more accurate.26
Screening for Potentially Lethal Conditions
With considerable media attention given to the recent sudden cardiac deaths of several prominent athletes, various medical factions have called for improving the detection of potentially lethal cardiovascular conditions.27,28 The most commonly debated methods of detection include the medical history, physical exam, ECG, and echocardiography (table 2). Genetic testing for HCM is not practical for clinical or screening use, with few exceptions.
In 1996, the American Heart Association (AHA) published a consensus statement consisting of 13 recommendations for screening competitive athletes for conditions that could lead to sudden death.27 The AHA writing group recommended cardiovascular screening for high school and college athletes and concluded that the history and physical exam were the "best available and most practical approach" to cardiovascular screening. Recommended screening questions included any previous exertional symptoms, such as chest pain, dyspnea, and syncope; a history of heart murmurs or hypertension; and a family history of cardiovascular conditions. Recommended items on the physical exam included blood pressure measurements, quality of pulses, presence of left ventricular outflow obstruction murmurs, and stigmata of Marfan syndrome.
Concern has been raised about the quality of current preparticipation cardiovascular screening in the United States.28-30 Glover and Maron28 surveyed 51 state high school athletic associations (including the District of Columbia), comparing their policies with the 13 recommendations from the 1996 AHA consensus statement. They found that of the 43 states that had history and physical exam forms, only 17 (40%) incorporated at least 9 of the 13 points from the AHA statement, and 12 (28%) incorporated only 4 or fewer items. Another survey conducted by Gomez et al,29 in which 193 of 254 high schools returned forms, found that 47 of the history and examination forms (24%) contained items regarding cardiovascular symptoms; 97 forms (50%) contained items regarding blood pressure and heart murmurs; and 57 forms (30%) contained items regarding family history. Only 32 forms (17%) contained items pertaining to all three areas. Pfister et al30 conducted a survey of 879 National Collegiate Athletic Association schools from Divisions 1, 2, and 3, comparing their history and exam forms with 12 items from the 1996 AHA consensus statement.27 Of the 625 forms available, only 163 (26%) contained from 9 to 12 of the AHA recommendations, and 150 (24%) contained 4 or fewer items.
The authors of these studies proposed that better standardization of the cardiovascular preparticipation exam across the nation could result in higher-quality screening and, therefore, enhanced chances of detecting potentially lethal cardiac conditions.28-30
One of the main problems with the AHA consensus statement recommendations is their lack of proven validity and reliability. Most of the recommended points in the history and physical exam have no epidemiologic data to prove their value, either alone or in combination, for predicting potentially lethal cardiovascular conditions. It is quite likely that their sensitivities and specificities are quite low. Many, if not most, of the conditions that lead to sudden cardiac death can often be asymptomatic and have minimal or no physical exam findings. Therefore, although no studies exist, if the AHA recommendations are applied to the large masses of young, competitive athletes across the nation, as advocated by Glover and Maron28 and Pfister et al,30 their positive predictive values could be exceedingly low, and the number needed to screen may be prohibitive.31
The recognized limitations of the history and physical exam as screening tools have prompted several investigators to evaluate echocardiography as a detection method for potentially lethal cardiac conditions.32-36 The rationale for echocardiographic screening is based on the premise that it is the best available means of detecting HCM, as well as other potentially lethal conditions. In five studies of echocardiography or ultrasonography performed in the preparticipation exam setting,32-36 no cases of HCM, Marfan syndrome, or any other potentially lethal cardiac condition were identified (table 3). All of the "positive" findings were incidental benign findings that conferred no risk of grave prognoses.
When examined in the context of what is known about HCM and other cardiac conditions, it is not surprising that screening echocardiography has such low yield. The actual prevalence of HCM in the general population is unknown but is estimated to be 1 in 500.37 Because many people who have HCM are symptomatic, some possibly severely so, the prevalence in the athletic population may be much lower, although actual data is unavailable. The prevalence of sudden cardiac death in young athletes is estimated to be between 1 in 200,000 and 1 in 300,000, and half of these deaths may be attributed to HCM.3-5 The total number of athletes screened in these echocardiographic studies32-36 is around 5,000, which may not come close to approximating the number needed to find any athletes at risk of sudden death. Certainly for every athlete who dies, several have conditions that place them at risk of dying; however, that number is uncertain and is probably very low.
Furthermore, the myocardial hypertrophy of HCM may not be detectable until growth is completed, and for some adolescent athletes who have a negative echocardiogram, hypertrophy could develop years later. No follow-up data are available from these studies. The actual sensitivity and specificity of echocardiography is not known,7 either for the detection of HCM or other potentially lethal conditions. Limited echocardiography, which is more feasible in mass screening efforts, may have substantially lower accuracy.
Another screening tool for cardiovascular conditions is the ECG. Conditions such as HCM often have characteristic ECG findings, and other conduction abnormalities can also be found by ECG. Fuller et al38 conducted preparticipation ECGs on 5,615 high school athletes. They found 146 initial abnormalities, mostly consisting of ST-T wave abnormalities, some left ventricular hypertrophy patterns, and other miscellaneous findings. Follow-up echocardiography on these 146 athletes yielded no abnormal findings. Sixteen of the 146 athletes who had abnormal ECGs had conduction abnormalities, and follow-up testing revealed 1 athlete with supraventricular tachycardia (SVT) who subsequently underwent ablation. In follow-up, the population had a single case of resuscitated ventricular fibrillation. This athlete had an anomalous coronary artery but had a normal screening ECG.
This study38 demonstrated no true positive findings, with the possible exception of the athlete with SVT who underwent ablation. Otherwise, the 146 "abnormal" ECGs were essentially all false-positives. Even though more than 5,000 athletes were screened, given the rarity of sudden death, this number is likely orders of magnitude too low to demonstrate positive findings. Although the authors conclude that ECGs have better sensitivities and specificities than the history and physical exam, their results demonstrate extremely low sensitivities and specificities for both ECGs and history with exams in the detection of potentially lethal cardiovascular conditions.
Ideal Screening CriteriaPossibly the most important characteristic of an effective screening program is that the suspected condition have a significant impact on public health. Sudden cardiac death in young competitive athletes fails to meet this criterion. Based solely on existing prevalence estimates, sudden death does not come close to approaching that of other screened, potentially lethal conditions such as breast cancer and prostate cancer. It could be argued that the relative youth and vitality of those at risk of sudden death place more weight on this condition compared with other lethal conditions that mainly affect older individuals. While this may be the case, the prevalence figures are nonetheless orders of magnitude different. Essentially, the prominent role of athletes in our society, and therefore their relative fame, is what has made sudden death such a high-profile public health issue.
A second characteristic of an ideal screening program is that the condition has an asymptomatic period during which detection is possible. Sudden cardiac death certainly meets this criterion.
Third, the outcome for a condition should be improved by treatment during the asymptomatic period. This criterion probably holds true for sudden cardiac death, although very little data exist. Deaths in patients who have these conditions certainly occur outside of competitive athletics, but the hemodynamic stresses from training and competition probably contribute to the risk of death. It is possible that restricting individuals from athletics only delays their inevitable deaths later. Some specific treatments exist for certain conditions, such as beta-blocker medications for HCM and Marfan syndrome, surgical correction of aortic root conditions in Marfan syndrome, or radiofrequency catheter ablation techniques and implantable defibrillators for some conduction problems. These treatments may also alter the natural course of the conditions but may not return athletes to competition.
Fourth, a screening test should be sufficiently sensitive to detect the disease during the asymptomatic period. The current available screening methods, individually and probably collectively (although never assessed collectively), fall well short of fulfilling this criterion. Although the AHA27 recommends a history and physical exam for screening, no data proves that this is a sensitive detection method. In fact, given the previously discussed nuances of the conditions likely to cause sudden death, the history and physical exam are probably very insensitive for detecting these conditions. Similar arguments apply to the low sensitivity of ECG and echocardiography.
Fifth, the screening test must be sufficiently specific to minimize false-positive test results. Current screening modalities fail to fulfill this criterion. The positive predictive value and the proportion of false-positive results partly depend on the prevalence of a condition within the screened population. For rare conditions, such as those that can lead to sudden cardiac death in young athletes, a test would need to be exceedingly specific to avoid large masses of false-positive results. This phenomenon has been clearly demonstrated in the echocardiography and ECG studies,32-36,38 in which essentially all positive results were false-positives. Neither history and physical exam, ECG, nor echocardiography came close to approaching necessary specificity levels. The high number of false-positive results leads to further procedures (and their costs), some possibly invasive, inappropriate lost time or elimination from athletics and recreation, and varying degrees of anxiety for patients and families.
The next criterion is that the screening methods should be acceptable to patients. No data exist on this matter, but screening, whether by history and physical exam, ECG, or echocardiography, while a minor nuisance to athletes, is otherwise acceptable. Most states and school systems require a general preparticipation sports physical. Therefore, adding a few items to the history and physical exam to cover the cardiovascular system, as recommended by the AHA, may not overburden examiners or athletes.
Whether preparticipation exams as a whole should be conducted has also been debated, but that is beyond the scope of this discussion. Certainly it is possible that more benign, nonlethal conditions may be uncovered during screening, and that can also be used as an argument for performing screening. The value, or lack thereof, of finding these other, more benign conditions is unclear and is also beyond the scope of this discussion.
Lastly, acceptable, relevant medical care should be available. Again, specific treatments, some more effective and some more acceptable than others, exist for many conditions that can lead to sudden cardiac death. Treatment for many of these conditions includes prohibition from athletics. Data on how acceptable this restriction is for most athletes do not exist.
Closely related to this last point is that patients be willing to comply with further workup and treatment. Again, no data exist, but it is likely that most athletes would reluctantly comply with further workup and treatment that includes restriction from competitive athletics. It is not known how many athletes who have been restricted from competitive athletics continue to train and participate in sport at a recreational level against medical advice and, therefore, still incur additional risk of sudden death.
The screening process for elite athletes in Italy is much more comprehensive and restrictive.25,39 Italian athletes undergo yearly physical exams, ECGs, stress testing, and echocardiography. Their vast volume of data suggest that this screening may reduce the risk of death from HCM, but extrapolation of these data to the US population is difficult.25,39 This type of national program incurs enormous costs, both financial and in numerous disqualifications from athletics.
Although the conduction of preparticipation exams is considered medically and legally necessary and benevolent by many, the actual utility of at least the cardiovascular component, specifically in terms of screening for lethal conditions, is questionable from an epidemiologic standpoint. While it would be difficult to recommend against screening, the current limitations should be recognized. The rarity of conditions that could lead to sudden cardiac death in young, competitive athletes makes screening a daunting task. Additionally, it is difficult to screen for most of these conditions because of their relative paucity of symptoms, signs, and test findings. Epidemiologic principles dictate that exceedingly high sensitivities and specificities of screening tests are necessary to have useful predictive values for rare conditions. None of the available screening methods (history and physical exam, ECG, or echocardiography) come close to approaching these degrees of sensitivity or specificity, and the number needed to screen is likely prohibitively high.
Dr Wen is a family practice physician in the department of family and community medicine at the University of Missouri-Columbia in Columbia, Missouri. Address correspondence to Dennis Y. Wen, MD, Department of Family and Community Medicine, University of Missouri-Columbia, M228 Medical Sciences Building, Columbia, MO 65212; e-mail to [email protected].
Disclosure information: Dr Wen 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.