Background The usefulness and modalities of cardiovascular screening in young athletes remain controversial, particularly concerning the role of 12-lead ECG. One of the reasons refers to the presumed false-positive ECGs requiring additional examinations and higher costs. Our study aimed to assess the total costs and yield of a preparticipation cardiovascular examination with ECG in young athletes in Switzerland.
Methods Athletes aged 14–35 years were examined according to the 2005 European Society of Cardiology (ESC) protocol. ECGs were interpreted based on the 2010 ESC-adapted recommendations. The costs of the overall screening programme until diagnosis were calculated according to Swiss medical rates.
Results A total of 1070 athletes were examined (75% men, 19.7±6.3 years) over a 15-month period. Among them, 67 (6.3%) required further examinations: 14 (1.3%) due to medical history, 15 (1.4%) due to physical examination and 42 (3.9%) because of abnormal ECG findings. A previously unknown cardiac abnormality was established in 11 athletes (1.0%). In four athletes (0.4%), the abnormality may potentially lead to sudden cardiac death and all of them were identified by ECG alone. The cost was 157 464 Swiss francs (CHF) for the overall programme, CHF147 per athlete and CHF14 315 per finding.
Conclusions Cardiovascular preparticipation examination in young athletes using modern and athlete-specific criteria for interpreting ECG is feasible in Switzerland at reasonable cost. ECG alone is used to detect all potentially lethal cardiac diseases. The results of our study support the inclusion of ECG in routine preparticipation screening.
- Cardiology prevention
Statistics from Altmetric.com
Sudden cardiac death (SCD) among young athletes is a rare (1–4/100 000 athletes/year)1–3 but dramatic event and has a profound psychological impact on the community, that regularly raises the question about a possible prevention. However, the usefulness and modalities of cardiovascular screening in young athletes is still a matter of dispute.4–7 The Italian experience suggested that a cardiovascular screening, including a resting 12-lead ECG, may prevent a substantial part of these tragic events,1 and the European Society of Cardiology (ESC)8 as well as major sports associations9 ,10 endorsed such a screening programme. On the other hand, the American Heart Association (AHA)11 recommends a screening protocol limited to medical history and physical examination (H&P). The role of ECG in the screening strategy is, therefore, one of the most controversial issue. ECG has been considered to have inherent limitations particularly related to the presumed high number of false positives which require additional cardiac examinations. Such aspect substantially raises the costs of a screening programme that should be implemented at large scale to prevent a low, albeit tragic, number of fatal events.11 It is, therefore, essential that specific criteria are used for athlete ECG interpretation. The aim of our study was to provide a cost and yield analysis of a preparticipation cardiovascular screening examination, including ECG in young athletes in Switzerland. The cost issue is, in fact, fundamentally important to discern whether to implement this policy at large scale.
This prospective observational study was performed in three Swiss public institutions: the Division of Cardiology of the San Giovanni Hospital (Bellinzona), the Center for Sports Medicine of the La Carità Hospital (Locarno) and the Division of Cardiology of the Civic Hospital (Lugano). The study was started in February 2011 and ended in April 2012. Since in Switzerland the preparticipation cardiovascular screening is not mandatory, athletes were self-referred and recruited through the sports clubs. The study conforms to the principles stated in the Declaration of Helsinki and was approved by the Scientific Ethics Committee of Canton Ticino.
Inclusion criteria were satisfied in case of competitive athletes (doing regular exercise training and participating in official athletic competitions)8 aged 14–35 years. If they already had a cardiovascular examination (for screening purpose or because of problems), more than 1 year should have elapsed before inclusion in the study. Exclusion criteria were athletes with known cardiovascular abnormalities. Written informed consent was signed by each athlete (for minors, a parent or a legal representative signed the consent).
All athletes were examined following the 2005 ESC consensus statement on cardiovascular preparticipation screening in young athletes. This protocol includes personal and family history, physical examination and ECG.8 Briefly, a positive family history was defined by the occurrence of premature sudden death (<50 years) among close relatives or in case of history of cardiomyopathies, Marfan syndrome, channelopathies, premature coronary artery disease (<50 years) or other potentially hereditary disabling cardiovascular diseases. A positive personal history was identified in case of exertional chest pain, excessive shortness of breath under effort, syncope of undetermined origin or palpitations. The physical examination was considered abnormal in case of features suggestive of Marfan syndrome, diminished femoral artery pulses, mid-systolic or end-systolic clicks, single or widely split and fixed second heart sound, heart murmurs (systolic ≥2/6 including auscultation in standing position and during Valsalva manoeuver, any diastolic), brachial blood pressure (both arms) in sitting position ≥140/90 mm Hg in >1 reading and irregular heart rhythm. The ECG was interpreted based on the 2010 ESC criteria12 readapted in the definition for specific ECG findings: QRS duration was considered abnormal if ≥120 ms13; T-wave abnormality was defined by a ≥1 mm inversion in ≥2 contiguous leads (minor T-wave inversion if ≥1 mm and <2 mm, deep negative T-wave if ≥2 mm)13 and by a T-wave inversion beyond V2 in ≥16-year-old adolescents,14 while in athletes of black ethnicity T-wave inversion in V1–V4 following convex ST elevation was not considered abnormal15; isolated voltage criteria for right ventricular hypertrophy were not considered abnormal16; QTc was considered pathological if ≥470 ms in males and ≥480 ms in females17 or ≤340 ms18; Brugada ECG pattern type 1 was considered pathological (in cases of Brugada type 2 or 3, an ECG was also recorded with V1 and V2 in the third and second intercostal space)19; ≥2 ventricular or atrial premature beats per ECG trace of 10 s were considered abnormal.13 All athletes were examined by cardiologists (in Bellinzona and Lugano) or sports physicians (in Locarno) with at least a 15-year clinical experience. The ECG was printed out as hard copy (25 mm/s, 1 mV/cm) and evaluated in consensus reading by two dedicated cardiologists (AM and MDV) with an extensive experience in evaluation of athletes, military staff and personnel engaged in high-risk occupations.
In order to exclude an underlying cardiopathy, all athletes with abnormal medical history, physical examination or ECG findings were proposed to undergo further examinations according to the recommendations for evaluating athletes with cardiovascular abnormalities.12 ,17 ,20 All indications for further examinations and the final diagnosis were consensually established by two cardiologists (AM and MDV).
The screening protocol (history, physical examination and ECG) was free of charge for all athletes enrolled while additional cardiac examinations due to abnormal findings at first-line screening were paid by the health insurances. For each athlete, we calculated the costs for the cardiovascular screening and for the examinations eventually performed to reach the final diagnosis, according to the current Swiss medical rates for public hospitals.21 Treatment and follow-up costs were not calculated. The costs of screening and additional second-line cardiac examinations were as follows (all in Swiss francs, CHF): history-taking (10–15 min) 33.30, cardiovascular clinical examination 42.55, ECG 26.95 (total for the entire screening examination 102.80), echocardiogram 312.08, exercise stress test 156.95, 24 h Holter monitoring 195.35, 24 h blood pressure monitoring 87.80, cardiac MRI 705.10, ECG with pharmacological exposure (adenosine) 155.75, signal-averaged ECG 69.25, genetic testing for long QT syndrome 3456, electrophysiological study and ablation of an accessory pathway 9695. The current exchange rate is about CHF1 for 0.81 and CHF1 for $US1.10. Notably, according to the Swiss medical system, the costs for the above-mentioned cardiac examinations include medical and non-medical personnel, infrastructures and equipment required to carry out the examinations.
Retrospectively, we also applied a hypothetical model of cardiovascular preparticipation screening including only H&P, as recommended by the AHA.11 For this purpose, we considered abnormal findings discovered, thanks to H&P. We then computed the additional examinations performed in these cases and we calculated the hypothetical costs of this programme. We do recognise the potential bias of such retrospective model, partially limited by the fact that the same two expert cardiologists (AM and MDV) examined most of the athletes (64%) and consensually established the indications for additional examinations and final diagnosis for all athletes.
Results are expressed as mean±SD. Statistical analysis was performed with SPSS V.19.
Table 1 shows the demographic characteristics of the study population. A total of 1070 consecutive athletes were examined and 75% of them were men. Almost all were of Caucasian origin. Several sports disciplines were represented, the most frequent of which were football and ice hockey. Most athletes competed at regional level. A minority of athletes (11%) previously underwent a cardiac examination, 8% for a screening purpose and 3% because of specific problems.
Abnormal findings and further cardiac evaluations
After the screening examination, a total of 67 athletes (6.3%) required further evaluations: 3.9% because an abnormal ECG, 1.4% due to physical examination and 1.3% due to medical history (selected athletes had more than one pathological element). Table 2 shows the abnormal findings and the tests chosen for each abnormality. In selected cases, several examinations were additionally carried out. Table 3 summarises second-line cardiac examinations actually performed and their relative costs.
Table 3 also shows the tests which would have been undertaken in the hypothetical case of a screening programme consisting only in H&P. In this case, a total of 28 athletes (2.6%) would have performed further examinations.
New cardiac diagnosis and clinical attitude
A new cardiac disease was detected in a total of 11 athletes (1.0%). Table 4 details their clinical characteristics including performed examinations, proposed therapy and decision about the eligibility for competition. Seven athletes were identified through positive history or abnormal physical examination (valvular diseases, systemic hypertension and atrial tachycardia). Four athletes (0.4%) had a cardiac disease potentially responsible for SCD (3 Wolff-Parkinson-White ECG-pattern and 1 long QT syndrome type 1 genetically proven). ECG alone identified all these athletes who would have, therefore, been missed in case of a hypothetical screening protocol including only H&P. All athletes with pre-excitation were asymptomatic and their treatment was individualised after a non-invasive evaluation17 ,22: an 18-year-old athlete underwent an electrophysiological study with ablation of the accessory pathway (refractory period 230 ms) while the other two were conservatively treated because they were considered to be at low risk (intermittent pre-excitation in one of them, 32-year-old body builder the other). The athlete with the genetically proven long QT syndrome 1 was asymptomatic and had a QTc of 485 ms. She was treated with β-blockers and restricted from sport participation. Of note, no cardiomyopathy was discovered (all athletes with negative T-waves had an echocardiogram and athletes with deep negative T-waves also underwent cardiac MRI).
Except for the one with long QT syndrome, all athletes were allowed to compete. We recommended an active follow-up for athletes with a new cardiac disease detected or with ECG abnormalities.
For 93.7% of athletes (those without abnormal findings at the screening examination), the cost only included the first-line screening examinations, that is, CHF102 80. The costs for the remaining athletes varied depending on the additional investigations performed (see also table 3). The highest cost was related to the athlete with Wolff-Parkinson-White ECG-pattern who underwent an electrophysiological study with ablation (CHF10 613). The cost was CHF157 464 for the overall screening programme, CHF147 per athlete and CHF14 315 per new cardiac finding.
In the hypothetical screening strategy consisting only in H&P, the cost would have been CHF91 855 for the overall programme, CHF85 per athlete and CHF13 122 per new finding.
In this study, we demonstrated that cardiovascular preparticipation examination with ECG in young athletes, carried out by experienced personnel and using modern and athlete-specific ECG criteria, is feasible in Switzerland at reasonable cost. Moreover, the low rate of abnormal ECGs in young athletes and the detectability of potentially lethal cardiac diseases with ECG alone confirm its major role in the screening protocol.
The prevention of SCD among young athletes is widely considered an important health problem raised by major sports and medical associations8–11; however, the modalities for its achievement still remain a matter of dispute.4–7 A screening programme intended to be implemented at large scale should include a large number of athletes to prevent a relatively low, albeit tragic, number of death. Hence, it should be sensitive, relatively cheap and able to detect disease early so that medical therapy can be implemented. Adding ECG to the screening programme, as recommended by the ESC and the major sports associations,8–10 has proven to be more effective in detecting cardiac diseases,4 ,6 particularly life-threatening conditions (cardiomyopathies, electrical cardiac disorders), often asymptomatic and without abnormal physical signs. Our study confirms once more that athletes with potentially lethal diseases were only identified on the basis of an abnormal ECG. Nevertheless, the AHA does not recommend ECG in the screening strategy.11 Indeed, one of the major controversies regarding the role of ECG is its presumed low specificity, resulting in a high number of false positives involving additional examinations and raising the costs of the screening programme.11
Regular physical activity induces structural and autonomic heart adaptations which, affecting ECG, may mimic ECG abnormalities seen in cardiovascular diseases. Therefore, ECG has been considered a non-specific tool for athlete cardiac evaluation. More recently, considerable progresses have been made in order to correctly interpret athlete ECGs. A first attempt was the 2005 ESC document8; however, based on these criteria, the rate of positive ECGs in different series still remained rather high, between 6% and 19%.23–26 The 2010 ESC recommendations represented a further advance in distinguishing physiological modifications due to exercise and abnormal findings suggesting cardiac diseases.12 Nevertheless, these criteria were in some aspects rather cautious (eg, QRS duration, abnormal T-waves, abnormal QTc, criteria for right ventricular hypertrophy and Brugada ECG pattern) or undefined (criteria for abnormal premature beats) so that they still resulted in a relatively high number of pathological ECGs varying from 6% to 11% in different series,12 ,27–29 up to 34% in another study.30 In our study, we adapted the 2010 ESC criteria in the aforementioned points according to literature data (see Methods section). After the beginning of our study, new criteria for the interpretation of the ECG of young athletes were published,31 and very recently, under the auspice of major international sports and cardiac societies, a group of experts edited the “Seattle Criteria” for interpreting athlete ECGs with thoughtful attention to balance sensitivity and specificity.32–35 These criteria are very similar to those adopted in our study. With 3.9% abnormal ECGs, we demonstrated that using modern and athlete-specific criteria, the rate of pathological ECGs in young athletes is low, as confirmed by the 4.7% of abnormal ECGs reported in another recent study.36 Therefore, few additional examinations were performed so that the entire screening programme of our study could be realised at the reasonable cost of CHF147 per athlete. If ECG would not have been part of the screening examination, as recommended by the AHA,11 the cost per athlete would have been cheaper (CHF85) but this strategy would have missed all potentially life-threatening cardiac abnormalities (4 of 11 new cardiac findings). On the other hand, the cost per new diagnosis would have been relatively similar (CHF14 315 vs CHF13 122 for the strategy of H&P+ECG, respectively H&P).
Literature data concerning the costs of a cardiac screening programme for young athletes are scarce and difficult to compare because of different methodologies and considerable variation in the prices of the healthcare system worldwide. In Italy,37 the estimated cost of a preparticipation screening examination with ECG is about CHF44 per athlete (CHF37 per screening examination, 10% of athletes requiring further evaluations at an estimated cost of CHF74 per athlete). However, while the Italian study just represents an estimated cost (rather conservative with current clinical standards), our study depicts the real world with current costs. In the USA,25 a study detailed the costs of a 5-year screening programme with ECG of 1473 college athletes, including follow-up tests and cost estimate for personnel and infrastructures: the cost was CHF552 per athlete and CHF62 578 per finding. In the case of screening with only H&P, the cost per athlete would have been CHF212 and cost per finding CHF62 495. As in our study, the cost per finding was similar in both strategies but the screening with ECG was more effective in discovering cardiac diseases. Of note, in that study, the ECG criteria adopted were the 2005 ESC ones8 with a 19% of abnormal ECGs requiring a greater number of additional examinations, thus explaining, in part, the higher programme costs. Recently, a study from Qatar36 showed that screening athletes with systematic echocardiography costs CHF453 per athlete and CHF73 794 per potentially life-threatening diagnosis, whereas the use of H&P+ECG strategy, as in our work, costs CHF241 per athlete and CHF39 194 per potentially life-threatening diagnosis.
Our study provided a cost analysis of a preparticipation cardiovascular screening including ECG and not the downstream costs of a screening programme, which includes the costs generated by screened individuals who become patients, costs of follow-up for athletes with abnormal findings, repeated screening tests, treatments and medications. Therefore, the true social cost of a screening programme is certainly higher than what we found in our study. Likewise, our study was not intended to demonstrate the cost-effectiveness of a screening programme.
The limited sample size of our study was insufficient to identify each of the rare screening-detectable diseases. Any patient identified with a disease, depending on the clinical circumstances, may significantly increase the total cost (for instance, an athlete requiring a defibrillator for primary prevention could increase the costs by 20%). Therefore, the presented cost may not be generalisable to a comprehensive screening programme.
The athletes in our study were self-referred for screening and it is conceivable that they may be less likely to have overt symptoms, artificially lowering the number of additional examinations and costs.
On the other hand, the physicians involved in our study were highly experienced and ECGs were interpreted in consensus reading by two dedicated cardiologists, largely fulfilling the recently published curriculum for the European Sports Cardiology qualification,38 so that the results of our study may not be generalisable.
If costs were the only criteria, screening programmes for athletes should be abolished because they are expensive and SCD are rare.39 Nevertheless, if a screening programme would be implemented, as widely suggested,8–11 then ECG should be part of it because it greatly improves its efficacy. ECG should be read with modern and athlete-specific criteria by experienced personnel in order to reduce false positives, as demonstrated in our study. In this way, the preparticipation cardiovascular examination can be realised at a sustainable cost, even where the health system costs are rather high as in Switzerland. The results of our study will, therefore, add arguments supporting ECG introduction into cardiovascular screening.
What are the new findings?
Using modern and selective criteria for interpreting ECG in young athletes, the rate of abnormal ECGs is low.
Cardiovascular screening with ECG in young athletes carried out by experienced personnel can be implemented with relatively few additional examinations.
The cardiovascular preparticipation examination is feasible in Switzerland at reasonable cost.
How might it impact on clinical practice in the near future?
ECG is a useful tool in the cardiovascular screening of young athletes.
Adding ECG to a preparticipation cardiovascular examination will help to detect more potentially life-threatening cardiac conditions.
In Switzerland, the inclusion of ECG in a pre-participation cardiovascular examination should not be postponed by cost concerns.
The authors are grateful to Alessandra Pia Porretta for her help in preparing the manuscript.
Contributors AM and MDV were responsible for the conception and design of the study. AM and MDV conducted the analyses, which were planned and checked with the coauthors AM wrote the first draft of the paper, which was critically revised by MDV, MW and AG. . All authors were involved in data collection over the study period. All authors contributed to interpretation of the findings and had full access to all data. The final version of the manuscript was approved by all authors. AG is the study guarantor.
Funding This study was supported by grants from the Swiss Heart Foundation, the Ente Ospedaliero Cantonale Ticinese, Medtronic Swiss and St Jude Medical Swiss.
Competing interests None.
Patient consent Obtained.
Ethics approval Scientific Ethics Committee of Canton Ticino.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.