Sudden death in high school athletes (HSA) on the playing fields is a rare, tragic, and alarming event. Formerly, two-thirds were a result of trauma, i.e., acute subdural hematoma and high level quadriplegia⁽³⁾. In the past 15 yr, nontraumatic deaths have outnumbered traumatic deaths by 2:1, primarily as a result of reduced frequency of the latter because of safer game rules and equipment⁽³⁾. Nontraumatic deaths have not been reduced⁽³⁾.

A conservative estimate is that 13 of the 2.7 million HSA per year in the United States die suddenly from a nontraumatic cause during sports participation ⁽³⁰⁾. Approximately 10 are cardiovascular deaths ⁽³⁰⁾. Hypertrophic cardiomyopathy (HCM) is the most common cause and is responsible for approximately 50% of these cardiovascular deaths ^(19,30). The remainder result from coronary artery anomaly, myocarditis, aortic stenosis, dilated cardiomyopathy, rupture of the aorta, right ventricular dysplasia, and primary conduction abnormalities^{(4,5,19,26,28-30)}.

The electrocardiogram (ECG) is a sensitive test for the detection of many of the cardiac diseases associated with sudden death in young athletes^{(2,4,5,22,27,32)}. For example, more than 95% of individuals with HCM, sudden death, and prior ECGs had left ventricular hypertrophy (LVH), ST-T wave abnormalities, and/or abnormal Q waves ⁽¹⁸⁾. Specificity of the ECG in preparticipation screening for causes of sudden cardiac death in athletes is felt to be relatively low ⁽²¹⁾. Data from large populations of young athletes comparing the specificity of the standard cardiovascular history and physical exam with ECGs are nonexistent. There are few studies on effective ways to screen large populations of young athletes for serious cardiovascular abnormalities ^(9,14-16,31).

The 1986 16th Bethesda Conference Cardiovascular Abnormalities in the Athlete: Recommendations Regarding Eligibility for Competition and the more recent 1994 26th Bethesda Conference Recommendations for Determining Eligibility for Competition in Athletes with Cardiovascular Abnormalities have outlined approaches for the athlete with known cardiovascular abnormalities ^(23,24). They also specifically list many ECG abnormalities, such as ventricular preexcitation, bundle branch block, premature ventricular contractions, and prolonged QT interval, for which they recommend further cardiovascular evaluation before approval for sports participation can occur.

In 1991 an 18-yr-old boy collapsed and died while playing high school basketball in our community. An autopsy revealed HCM. He underwent a preparticipation history and physical examination with normal results 3 months before collapse; an ECG had not been obtained. This unfortunate event provoked a community outcry for better screening and resulted in the current study. Realizing that an ECG in this young athlete probably would have been abnormal, we elected to determine whether the addition of an ECG to the standard preparticipation history and physical examination would improve the detection of serious or potentially serious cardiac abnormalities and to compare the specificity (number of false-positives) of both screening approaches.

SUBJECTS AND METHODS

Subjects. In conjunction with the approval of school officials and coaches, male and female student athletes at 30 selected high schools in Northern Nevada were invited to undergo, without charge, preparticipation screening. Schools were chosen based on their geographic proximity to urban and rural cardiology clinics of Sierra Nevada Cardiology Associates.

All invited schools participated. Student participation was supported, encouraged, and coordinated by school coaches. The exact percentage of student participation is not known, although it was estimated to be greater than 95% by school coaches. Informed consent was obtained from the students and their parents.

Screening procedures. Screening consisted of a cardiac history, cardiovascular auscultation/inspection, blood pressure measurement, and an ECG. Tests were performed on all HSA in groups of 50-300 athletes. History, blood pressure measurement, and the ECG (12-lead without rhythm strip) were obtained by a cardiac technician. Individual histories of cardiovascular medications were not obtained. Cardiovascular auscultation/inspection and interpretation of the ECGs were performed by Board eligible/certified cardiologists. Blood pressure measurements were obtained in both arms in the sitting position and repeated in 10 min if > 140/90 mm Hg. If indicated, color flow Doppler and two-dimensional echocardiograms were performed by experienced cardiac sonographers.

Screening abnormalities of the cardiac history. Nonvasovagal and nonexercise-related syncope, exercise-related symptoms (angina, excessive dyspnea, or syncope during or shortly after exercise), frequent dizziness or palpitations, history of a heart murmur, and family history (first degree relatives) of HCM or sudden cardiac death before age 35 yr by Marfan's syndrome.

Screening abnormalities by cardiovascular auscultation and inspection. Systolic murmur ≥ grade II/VI or increase in intensity from supine to standing, any diastolic murmur, abnormal second heart sound, or Marfanoid appearance.

Screening abnormalities of blood pressure. Moderate elevation was defined as systolic blood pressure of 150-159 mm Hg and/or a diastolic pressure of 95-99 mm Hg. Severe elevation was defined as systolic pressure≥ 160 mm Hg and/or a diastolic pressure ≥ 100 mm Hg⁽²⁴⁾. Moderate or severe elevation in blood pressure was considered an abnormality.

Screening abnormalities of the ECG. Bundle branch block, premature ventricular beats (≥1), supraventricular tachycardia, atrial tachycardia, nonparoxysmal atrioventricular junctional tachycardia, atrial flutter, atrial fibrillation, ventricular preexcitation, ventricular tachycardia, second and third-degree atrioventricular block and prolonged QT interval are findings that require, per the 16th Bethesda Conference, further testing before approval for sports can be considered. Q waves, ST-T wave abnormalities, and LVH were considered abnormal primarily as screening for HCM. LVH was present in males if the voltage of SV1 + RV5 ≥ 50 mm and, in females, if the voltage of SV1 + RV5 ≥ 40 mm ⁽¹⁾. ST wave abnormalities were present if the ST segment deviated from baseline by≥ 1 mm in two or more leads. Symmetric T-wave inversion or flattened T-waves were abnormal if present in any two leads except for V1-3. A Q wave was abnormal if it was ≥ 0.04 s, ≥ 3 mm deep, and present in two or more leads excluding leads V1-2 ⁽¹⁶⁾. Ventricular preexcitation was present if the PR interval was < 0.12 s and the QRS was> 0.09 s and had a delta wave ⁽¹²⁾.

Post-screening procedures. On site M-mode, 2-D, and color flow Doppler/echocardiography were performed if screening abnormalities were found; we were looking primarily for HCM, dilated aorta, dilated cardiomyopathy, or aortic stenosis. Treadmill stress testing was performed only if there was a history of angina and/or exercise-induced syncope, primarily to detect evidence of ischemia reflective of a possible coronary artery anomaly.

Abnormalities of the echocardiograms and exercise treadmill tests. Abnormalities were based on the criteria of the American Society of Echocardiography ⁽⁸⁾. LVH was present if the thickness of the ventricular septum or posterior wall was > 11 mm⁽⁸⁾. The diagnosis of HCM consisted of maximal diastolic left ventricular wall thickness of ≥ 15 mm and was based on the 16th Bethesda Conference Recommendations ⁽²³⁾.

The Bruce protocol for treadmill exercise stress testing (symptom limited) was used; the test was abnormal if ≥ 1 mm horizontal ST segment depression in one or more leads occurred during exercise ⁽⁶⁾.

Outcome measures. The 16th Bethesda Conference lists certain cardiovascular abnormalities (primary conduction abnormalities, arrhythmias, severe hypertension, congenital and acquired valvular heart disease, cardiomyopathies, and coronary artery anomalies) which, if detected in the athlete, either preclude certain competitive athletic activity or require further evaluation before consideration for sports approval can be given⁽²³⁾. These abnormalities are believed in some cases to either place the athlete at risk for sudden cardiac death or for disease progression ⁽²³⁾. The detection of any of these diseases or abnormalities from an abnormal screening test was considered an outcome measure.

Approval for sports participation and follow-up. HSA were approved for sports participation if there were no screening abnormalities by history, auscultation/inspection, blood pressure measurement, and ECG. HSA with a screening abnormality by history or cardiovascular auscultation/inspection were not approved unless the echocardiogram and/or treadmill test was normal. Athletes with severe elevation of blood pressure were not approved, were considered to have an outcome measure, and were referred to their family physician for follow-up. Athletes with moderate blood pressure elevation were approved if their echocardiograms were normal. HSA with LVH or abnormalities of Q waves, ST segments, or T-waves on ECG were approved if the echocardiogram was normal, thereby ruling out HCM. HSA with other ECG screening abnormalities (certain arrhythmias and specific primary conduction disturbances) were not approved for athletics, were considered to have an outcome measure, and were referred to their primary care physician for further cardiovascular evaluation as recommended by the 16th Bethesda Conference ⁽²³⁾. Likewise, HSA with abnormal post-screening echocardiograms were managed as recommended by the 16th Bethesda Conference. Athletes with a positive post-screening treadmill test were referred to their primary care physician for consideration for stress thallium imaging to further evaluate for an anomalous coronary artery. HSA with a screening abnormality by history, cardiovascular auscultation/inspection, or ECG (specifically LVH, abnormal Q-waves, and/or ST-T wave abnormalities) whose echocardiogram and/or treadmill test were normal were defined as having a false-positive screening abnormality. HSA with a screening abnormality whose echocardiogram had a minor abnormality which per the 16th Bethesda Conference did not preclude athletic activity or require further testing were defined as having a false-positive screening abnormality.

RESULTS

Screening abnormalities. Of the 5,615 HSA (3375 males, 2240 females; ages 13-19 yr) screened, 5,033 (90%) had no screening abnormality by history, cardiovascular auscultation, blood pressure measurement or ECG, and were approved for participation in high school sports. Five hundred eighty-two(10%) HSA had screening abnormalities (history in 115 (2%), cardiovascular auscultation/inspection in 175 (3.2%), blood pressure measurement in 20(0.3%), ECG in 146 (2.5%). Combined abnormalities were detected in 126 (2%). Of the 115 with an abnormal history, 42% had exercise symptoms, 30% history of a murmur, 10% frequent dizziness, 10% nonvasovagal fainting, 5% frequent palpitations, and 3% a family history of early sudden cardiac death. Of the 175 with abnormal cardiovascular auscultation/inspection, 70% had a grade II/VI systolic murmur, 18% had a grade III/VI systolic murmur, 8% had an abnormality of S2, 3% had a diastolic murmur, and 1% had a murmur louder with standing. One student had severe aortic regurgitation by cardiovascular auscultation. Of the 20 with a blood pressure abnormality, 75% had moderate elevation and 25% had severe elevation. Of the 146 with an abnormal ECG, 65% had ST wave and/or T wave abnormalities, 21% LVH, and 4% abnormal Q waves. Ten percent had either premature ventricular beats, ventricular preexcitation, right bundle branch block, or supraventricular tachycardia.

Post-screening echocardiogram and treadmill test results. Echocardiograms, performed in 582 (10%) for abnormal screening results, were normal in 538 and showed only minor abnormalities in 43. Minor abnormalities consisted of mild tricuspid regurgitation in 17, mild mitral regurgitation in 16, a bicuspid aortic valve in four, minimal LVH (wall thickness = 12 mm) in three, mild aortic regurgitation in one, mild pulmonic regurgitation in one, and mild pulmonic stenosis in one. Per the 16th Bethesda Conference, these HSA with minor echocardiographic abnormalities did not require further testing and were approved for sports activity. The echocardiogram confirmed the presence of severe aortic regurgitation in one HSA whose ascending aorta was not dilated. This HSA per the 16th Bethesda Conference was not approved for athletics. This auscultatory finding was an outcome measure.

Of the 146 echocardiograms performed for abnormal ECGs, 130 were performed for LVH, ST-T wave changes, or Q-waves predominantly to exclude the most common cause of sudden cardiac death in HSA, HCM. Of these 130 HSA (2.3% of HSA screened), none were found to have HCM or any other serious cardiac abnormality. Their ECG screening abnormalities were falsepositive findings. Only one of 28 HSA with LVH by ECG had minimal LVH by the echocardiogram. All three HSA with minimal LVH by echocardiogram had normal blood pressure measurements and were judged to have an athletic heart⁽¹³⁾. Echocardiograms were performed on 16 HSA (0.3% of HSA screened) with primary conduction abnormalities or arrhythmias: premature ventricular beats, ventricular preexcitation, right bundle branch block, and supraventricular tachycardia. All echocardiograms were normal. Per protocol and as recommended by the 16th Bethesda conference for these specific abnormalities, additional testing was required before approval for sports activity. Their screening abnormalities were considered false-positive findings. No HSA had exercise-induced syncope, the other indication for treadmill testing.

Approval for sports participation following post-procedure testing(Fig. 1). Of the 5,615 athletes tested, 5,593(99.6%) HSA were approved for participation in high school sports. Twenty-two(0.4%) HSA (18 males, 4 females) were not approved for participation and per protocol and, following the recommendations of the 16th Bethesda Conference, were referred for further cardiovascular testing/evaluation. Cardiac history did not lead to the detection of any outcome measures. Cardiovascular auscultation/inspection led to the detection of one HSA (1/6000) with severe aortic insufficiency who was not approved for athletics and was referred for treatment. Five (1/1000) HSA were not approved owing to severe hypertension. Sixteen (1/350) HSA were not approved because of an abnormal ECG. Six had ventricular preexcitation, five premature ventricular beats, four right bundle branch block, and one supraventricular tachycardia. They were referred for further testing as recommended in the 16th Bethesda Conference. There were no cases of atrial flutter, atrial fibrillation, atrial tachycardia, nonparoxysmal atrioventricular junctional tachycardia, ventricular tachycardia, second and third degree atrioventricular block, left bundle branch block, or QT interval prolongation.

Follow-up. During the 3-yr study period (1991-1994), no HSA sustained sudden death during sports participation. One HSA developed ventricular fibrillation during track practice and was successfully resuscitated. Coronary angiography revealed an anomalous right coronary artery. Its point of origin was adjacent to the ostium of the left coronary artery; its path was between the aorta and pulmonary artery. This student had a normal screening examination. Treatment was successful coronary artery bypass surgery.

Results of further testing as recommended by the 16th Bethesda Conference are available in one HSA with supraventricular tachycardia. He underwent radiofrequency ablation of an accessory pathway and returned to sports activity. We do not have the results of further testing performed on the other 15 HSA with primary conduction abnormalities/arrhythmias. The five HSA with severe hypertension and one HSA with severe aortic insufficiency, all denied approval for sports participation, did not require further testing but underwent treatment for their serious cardiovascular conditions.

DISCUSSION

Recent recommendations state that the purpose of preparticipation cardiovascular screening is identification of pre-existing conditions, which in the presence of intense athletic training, increase the risk of sudden cardiac death (or disease progression) ⁽²¹⁾. These recommendations advise that preparticipation screening should include a complete cardiovascular history and physical examination without the use of noninvasive testing ⁽²¹⁾. A 12-lead ECG is acknowledged to add to the diagnostic power of the complete cardiovascular history and physical examination but is assumed in large population screening to be impractical and have relatively low specificity ⁽²¹⁾. Do the data we and others have recently generated on this topic support these recommendations? Or do these data suggest a screening ECG is a reasonably specific, practical, and sensitive test for the causes of sudden cardiac death in young athletes?

Our study of nearly 6,000 HSA documented the much greater effectiveness of preparticipation ECGs than history and auscultation/inspection in detecting cardiovascular abnormalities requiring further testing before sports approval could be given according to the 16th Bethesda Conference. It did not have the power to evaluate the effectiveness of various screening modalities for the detection of definite causes of sudden cardiac death in HSA because of the exceedingly low prevalence of these abnormalities in young athletes. Epstein and Maron ⁽⁷⁾ have suggested that for every one cardiac death in a young athlete there are 10 young athletes with disease capable of causing sudden cardiac death. It is very conservatively estimated that there are 10 sudden cardiac deaths per year in HSA in the United States; one-half result from HCM ^(19,30). Further, there are approximately 2.1 million HSA per year involved in specific high school sports in which sudden death is reported ⁽³⁰⁾. On this basis, we estimate that the prevalence of HCM in HSA is only 50 in a total of 2.1 million HSA or 1/40,000, and the overall prevalence of cardiovascular abnormalities capable of causing sudden cardiac death in HSA is 100 in a total of 2.1 million HSA or 1/20,000. The presence of HCM in the general population is 1/500 ⁽¹⁷⁾. It apparently is much less in young athletes probably because of preselection. Not surprisingly our evaluation of almost 6000 young athletes did not detect any HSA with HCM or other causes of sudden cardiac death, most likely owing to their exceedingly low prevalence in young athletes. Two large screening studies of 4,567 HSA using echocardiography likewise found no HSA with HCM or other common causes of sudden cardiac death ^(9,31). It would require a study of, we estimate, 120,000 HSA to evaluate the effectiveness of screening. Six HSA would be predicted to have conditions predisposing to sudden cardiac death. Three of those would have HCM.

A recently published profile ⁽²⁰⁾ of 115 young American athletes with sudden cardiac death and prior standard preparticipation history and physical examination does provide data on the sensitivity of screening. In only 3% of those screened was there any suspicion of cardiac disease, and in only 1% was the correct diagnosis made. Forty-eight of those with sudden cardiac death were found to have HCM as the etiology of their death. In only one did the standard history and physical examination detect the serious underlying cardiac condition. The sensitivity of the standard cardiac preparticipation screening, which includes a complete cardiac history and physical examination, is at most 3% for detection of cardiac abnormalities which can lead to sudden cardiac death. In contrast, the ECG is abnormal in 95% of HSA with HCM and sudden cardiac death, in nearly all patients with prolonged QT interval or ventricular preexcitation, and in many patients with myocarditis, coronary artery anomalies, right ventricular dysplasia, dilated cardiomyopathy and aortic stenosis^{(2,4,5,11,18,22,25,32)}. Our estimate of the overall sensitivity of the ECG for detection of the common causes of sudden cardiac death in young athletes is between 60% and 70%.

The screening of large populations such as in HSA preparticipation examinations can be very problematic because of the cost incurred in evaluation of false-positive responses. It is recommended that the cardiovascular history include key questions of: exercise associated symptoms of angina, excessive dyspnea, or syncope; family history of premature cardiovascular death or disability before age 50 yr; or specific knowledge of certain cardiovascular conditions (HCM, dilated cardiomyopathy, long QT interval, Marfan's syndrome, or clinically important arrhythmias)⁽²¹⁾. The frequency of a positive response to these specific questions in our study cohort was 1.5%. With an estimated prevalence of 1/20,000 and a sensitivity of 3%, 660,000 HSA would need to be screened by cardiovascular history to detect one HSA of the predicted 33 HSA with serious disease (Table 1). Ten thousand HSA would have false-positive responses requiring cardiovascular referral and noninvasive testing. More broad-based questions, as were used in our study, reduce the specificity without improving sensitivity. We conclude that the current recommended cardiovascular screening history is almost fruitless and is relatively expensive because of the evaluation of false-positive responses.

Screening auscultation/inspection detected one HSA (1,6000) in our study with a serious cardiovascular abnormality (severe aortic regurgitation), not commonly associated with sudden cardiac death but associated with disease progression during certain intense physical activities. Not unexpected owing to the low prevalence of abnormalities, no HSA evaluated for ≥ grade II/VI systolic murmur had a serious cardiovascular abnormality. We conclude that false-positive responses can be minimized without compromising sensitivity by limiting the cardiovascular preparticipation examination to: ≥ grade III/VI systolic murmur or a systolic murmur increasing in intensity from supine to standing position, any diastolic murmur, or a finding of marfanoid appearance. This approach is similar to the recent recommendations for a standard cardiovascular physical examination ⁽²¹⁾. The frequency of a positive response to the specific physical findings in our study cohort was 0.7%. Of 660,000 HSA screened in this way, one of the predicted 33 HSA with serious disease would be detected, and 110 HSA would be diagnosed as having a significant cardiovascular abnormality which could worsen with intensive training (Table 1). Approximately 5,000 HSA would have a false-positive finding and need referral for echocardiography. Preparticipation auscultation/inspection appears more fruitful than preparticipation history and is less costly because fewer HSA require evaluation of false-positive responses. It is, in contrast to preparticipation history, labor intensive requiring a qualified physician/healthcare worker to perform the examination.

Blood pressure measurement is recommended to be included in the preparticipation examination ⁽²¹⁾. Severe blood pressure elevation with certain types of intense physical activity may result in disease progression ⁽²⁴⁾. Of 660,000 HSA screened, 660 would be found to have severe hypertension (Table 1) and would not be approved for athletics. This effort is easily performed by a healthcare worker, relatively fruitful, and not costly in terms of evaluating false-positive responses.

ECG evidence of LVH, ST-T wave abnormalities, and Q-waves produced false-positive responses in 2.3% of HSA screened. Of 660,000 HSA with a preparticipation ECG, 20 of the predicted 33 HSA with disease predisposing to sudden cardiac death will be detected and 15,000 HSA will need referral for echocardiography because of a false-positive ECG (Table 1). ECG findings of primary conduction abnormalities and arrhythmias as defined by the 16th Bethesda Conference are found in 0.3% of HSA screened. Screening ECG in our study detected one HSA (1/6,000 HSA) with a serious arrhythmia (chronic supraventricular tachycardia), not known to be associated with sudden cardiac death but possibly associated with disease progression to left ventricular dysfunction ⁽²³⁾. Between 5% and 10% of the cases of sudden cardiac death of young athletes are presumed to have a primary arrhythmic etiology (prolonged QT interval, ventricular preexcitation, etc.) ^(5,30). Therefore, the prevalence is approximately one in 200,000 HSA. Of 660,000 HSA screened for these findings by ECG, we feel all three HSA with a primary arrhythmic/conduction problem at risk for sudden cardiac death would be detected and 110 HSA with a serious arrhythmia at risk of disease progression would be identified. Two-thousand HSA would be referred for cardiovascular consultation and further testing because of a false-positive ECG (Table 1).

Overall, ECG screening for all of the common causes of sudden cardiac death in young athletes has a sensitivity, we estimate, of 60-70% and a specificity of 97.4% with a false-positive response rate of 2.6%. Of 660,000 HSA screened by ECG, 23 of the predicted 33 HSA with serious disease will be detected and 17,000 HSA will be referred for further testing. Complete cardiovascular history and auscultation/inspection has a sensitivity of 3%, and a specificity of 97.8% with a false-positive response rate of 2.2%. Of 660,000 HSA screened by standard history and auscultation/inspection, one of 33 HSA with serious disease will be detected and 15,000 HSA will be referred for further study. In comparison, ECG screening is clearly more sensitive and has similar specificity.

We have established the practicality of on-site ECG performance in groups of 50-300 HSA, each performed by a cardiac technician not necessarily with an M.D. present. The ECGs can be computer interpreted and overread by a qualified physician off-site. The cost can be minimized through competitive bidding for the service. We estimate a cost of $10 per ECG including technical and professional fees. The feasibility and cost of ECG need to be compared to the same measures for performance of a history and physical examination, most often done by a qualified physician/healthcare worker. In an ongoing, nonvolunteer setting involving preparticipation screening of large populations of young athletes, the cost of performing a cardiac history/physical examination and ECG may be similar. Further, we recommend an ECG be performed only once in each athlete's high school career.

Preparticipation examination of HSA is required by the majority of states⁽¹⁰⁾. The requirements and intent of these examinations vary considerably from state to state ⁽¹⁰⁾. If legislators continue to mandate through state agencies preparticipation examinations, it behooves them to ask what the objectives are and whether such objectives are being met with sensitive, practical, specific, and cost-effective screening tools.

If the objective is to reduce the number of nontraumatic sudden deaths on the high school playing fields, then we suggest performing an ECG once in the athletic career of an HSA engaged in sports in which sudden death has been reported.