Visual or computer-based measurements: Which is important for the interpretation of an athlete's electrocardiography?

SUMMARY

OBJECTIVE:

Preparticipation screening of athletes by electrocardiography is the most crucial step in determining sudden cardiac death risk factors. Several electrocardiography interpretation software programs have been developed for physicians practicing in this field. Our study aimed to assess cardiopoint sudden death screening module by comparing its findings with two cardiologists using Seattle and International criteria.

METHODS:

A total of 303 licensed national athletes (37% females) were enrolled. electrocardiographies were examined by the cardiopoint sudden death screening module using Seattle criteria and cardiologists. The consistency between cardiologists and software was compared, and the confidence assessment of the module was tested.

RESULTS:

With regard to Seattle criteria, moderate consistency was found between the cardiopoint sudden death screening module and the 1st (κ=0.41) and 2nd cardiologist (κ=0.59). Consistency between two cardiologists was moderate (κ=0.55). When we applied International criteria, there was moderate consistency between the module and the 1st cardiologist (κ=0.42), and good consistency between the module and the 2nd cardiologist (κ=0.63). Consistency between the two cardiologists was good (κ=0.62).

CONCLUSION:

The cardiopoint sudden death screening module had similar agreement with cardiologists based on both criteria. However, the software needs to be updated according to International criteria. Using computer-based measurements for preparticipation screening will help to save time and provide standardization of electrocardiography interpretation.

KEYWORDS:
Athlete; Electrocardiography; Exercise; Cardiac sudden death

INTRODUCTION

Sudden cardiac death (SCD) is one of the leading causes of death in sports participants¹,². Preparticipation screening, consisting of medical history, physical examination, and a resting 12-lead ECG, aims to identify pre-existing cardiovascular abnormalities that may lead to SCD³.

ECG interpretation criteria, such as the European Society of Cardiology (ESC) criteria, Seattle criteria, and International criteria, have been developed for preparticipation screening⁴–⁷. These criteria have clearly and practically delineated normal, borderline, and pathological ECG findings in athletes aged between 12 and 35 years⁷. Each revision of ECG criteria resulted in improved specificity without compromising sensitivity.

Automated ECG interpretation is fast and time-saving; however misdiagnosis is also possible. One such software program is the cardiopoint sudden death screening (SDS) module⁸. This module examines ECGs using Seattle criteria. Slaby et al. reported that the cardiopoint SDS module has a high negative predictive value with variable levels of sensitivity and specificity⁹.

The aim of our study was to compare the ECG interpretation of the cardiopoint SDS module to that of two cardiologists in the preparticipation screening of sports participants.

METHODS

Participants

We enrolled 303 licensed national athletes (37% females) from 34 sports disciplines between the ages of 13 and 35 years from October 1, 2017, to April 1, 2018. All athletes underwent cardiovascular screening, including medical histories and physical examination. Morphometric and demographic data were also obtained.

Ethical approval and consent

Athletes and, if needed, their parents gave informed consent to participate in the study. This study complied with the Declaration of Helsinki, and the research protocol was approved by the local institutional ethics committee.

Electrocardiography

We used a BTL^® 08 MT Plus (BTL, United Kingdom) 12-lead ECG tool. All ECGs were automatically sent to the cardiopoint SDS module via Wi-Fi LAN network and were automatically interpreted using Seattle criteria. ECG findings were also analyzed by two cardiologists with different levels of expertise using both Seattle and International criteria.

Statistical analysis

All statistical analyses were performed using SPSS Statistics for Windows, version 22 (IBM Corporation, Armonk, NY). Continuous variables are presented as means±SD and categorical variables are presented as percentages. Values of p<0.05 were considered statistically significant. Cohen's kappa (κ) statistics were used to determine the consistency between the cardiopoint SDS module and observers. κ (kappa) scores between 0.01 and 0.20 were classified as none to slight, 0.21 and 0.40 as fair, 0.41 and 0.60 as moderate, 0.61 and 0.80 as good, and 0.81 and 1.00 as almost perfect agreement.

RESULTS

The baseline demographics of athletes are shown in Table 1.

Table 1
Baseline demographics.

Variables	n=303
Age (years)	18.7±4.1
Height (cm)	173.2±10.0
Weight (kg)	66.9±14.0
Body mass index (kg/m2)	22.1±3.3
Heart rate (beats/min)	75±11.6

Findings based on Seattle criteria

The cardiopoint SDS module detected 22 (7.3%) ECGs as abnormal. Both cardiologists found 14 (4.6%) ECGs as abnormal. There was moderate consistency between the cardiopoint SDS module and the first cardiologist (κ=0.41), as well as the second cardiologist (κ=0.59). Furthermore, moderate consistency was found between the two cardiologists (κ=0.55) (see Figure 1).

Figure 1
Consistency levels of the cardiopoint sudden death screening module and the 1st and 2nd cardiologists in the evaluation of the Seattle and International criteria.

Findings based on International criteria

The ECG findings of the cardiopoint SDS module were also re-evaluated using International criteria. Further examination was suggested in 14 (4.6%) athletes. A total of 7 (2.3%) athletes were suggested for further examination by the first cardiologist, and 12 (4%) athletes were suggested by the second cardiologist. There was moderate consistency between the cardiopoint SDS module and the first cardiologist (κ=0.42), while the kappa statistic between the module and the second cardiologist showed higher consistency (κ=0.63). Likewise, there was good consistency between the two cardiologists (κ=0.62) (see Figure 1).

Electrocardiographic findings

The ECG parameters for which the cardiopoint SDS module had high sensitivity and specificity, as well as the parameters for which the module had low sensitivity, are shown in Figure 2.

Figure 2
Consistency levels of the cardiopoint sudden death screening module and the 1st and 2nd cardiologists in the evaluation of electrocardiography findings.

The cardiologists detected 67 early repolarizations and 2 T wave inversions (TWIs); however, these findings were not detected by the module. The module reported one early repolarization and one TWI, which were defined as false positives by the cardiologists. The cardiopoint SDS module found five ECGs with ST segment depression. However, both cardiologists defined these ECG changes as normal findings occurring secondary to right bundle branch block (RBBB). Furthermore, five complete RBBB were detected by the cardiologists; however, none of these were defined by the cardiopoint SDS module.

The cardiopoint SDS module did not report the parameters that were not defined for this software: respiratory sinus arrhythmia and juvenile TWI. The cardiologists found 4 sports participants had juvenile TWI and 29 sports participants had respiratory sinus arrhythmia.

DISCUSSION

In this study, the cardiopoint SDS module provided ECG interpretation results similar to cardiologists. Furthermore, when we applied International criteria to the cardiopoint SDS module, the consistency between the module and cardiologists becomes higher.

In preparticipation screening of athletes, strong correlations have been observed between the clinicians and automatic ECG analysis using the Cardea software program with the detailed descriptions of ECG findings¹⁰. Hyde et al. explained that the difference in pathologic Q wave definition between Seattle criteria and International criteria decreased false positive rates¹⁰. A new and detailed definition of ECG findings may allow automatic ECG interpretation devices to give more consistent results.

One of the major differences between the Seattle and the International criteria is that some abnormal findings based on the Seattle criteria shifted to the borderline class in the International criteria. Left axis deviation, left atrial enlargement, and complete RBBB were defined as abnormal findings based on the Seattle criteria, while all of these appeared in the borderline class in the International criteria⁶,⁷.

Although previous studies showed no relationship between these abnormal findings in ECG and morphological changes of left or right heart structures¹¹, these findings may still be abnormal ECG findings as assessed by more sensitive imaging modalities including cardiac magnetic resonance imaging (MRI) due to showing tissue characterization. Therefore, if two or more borderline ECG findings are seen on the surface ECG, it can be useful to lead further evaluation and close follow-up⁷. The other possible factor supporting our findings was that the age group in which TWI was considered as an abnormal ECG finding. In International criteria, TWI is accepted as a normal ECG finding in athletes below 16 years of age whereas TWI is considered an abnormal ECG finding irrespective of age according to Seattle criteria⁶,⁷. In our study, athletes below 16 years of age comprised 33% of the study population. So, we can conclude that using International criteria instead of Seattle criteria reduces the rate of abnormal ECG findings.

Importantly, the cardiopoint SDS module had a high sensitivity and specificity for the correct calculation of corrected QT (QTc). Long or short QT is associated with an increased risk of fatal ventricular arrhythmias in young sports participants¹²,¹³, and unfortunately, the QT interval cannot be accurately calculated by clinicians, including experienced cardiologists¹⁴. Therefore, the evaluation of ECG by clinicians and software together can give more reliable results.

Beyond this accuracy, our study showed that the cardiopoint SDS module did not have the same power as that in a retrospective study by Slaby et al.⁹ for the determination of pathological Q wave and left axis deviation. Pathological Q wave and left axis deviation were seen commonly in hypertrophic cardiomyopathy (HCM) subjects¹⁵–¹⁷. HCM is one of the leading causes of SCD in sport participants in the USA¹⁸. Additionally, silent myocardial infarction can lead to the development of Q wave¹⁹. Moreover, the software has difficulty identifying myocardial infarction related ECG findings²⁰. When all these findings are evaluated together, the cardiopoint SDS module may be insufficient in subjects with a high risk of developing SCD such as HCM. This finding once again demonstrates the importance of ECG analysis with clinicians.

The study by Hyde et al. showed that the Cardea software program had a trend of being more useful in clinical practice with the technological development and standardization of measurements¹⁰. With more clearly defined ECG parameters, each update of ECG interpretation criteria provided improvement in specificity and decreased false positive results compared to the previous criteria⁷. These findings coupled with the fact that using the International criteria may lead to better consistency between the clinicians in preparticipation screening for young athletes. More consistent results may provide a reduced risk of SCD during exercise and unnecessary disqualification of athletes who do not have cardiovascular disease. In the future, the cardiopoint SDS module may be the recommended software to analyze ECG in sports participants.

LIMITATIONS

Our study included only a small number of athletes. Additional diagnostic tests have not been performed in athletes with ECG changes. Another limitation is that in the evaluation of software-physician consistency, physicians should not be compared with a standard ECG device other than the cardiopoint SDS module.

CONCLUSION

Modern ECG interpretation software analyzes ECGs in a short time with high accuracy. In particular, the cardiopoint SDS module coupled with International criteria may provide more consistent results, and its clinical use may help provide the standardization preparticipation screening for the determination of SCD risk in sports participants.

• Funding: none.

REFERENCES

• ¹
  Harmon KG, Asif IM, Klossner D, Drezner JA. Incidence of sudden cardiac death in National Collegiate Athletic Association athletes. Circulation. 2011;123(15):1594-600. https://doi.org/10.1161/CIRCULATIONAHA.110.004622
  » https://doi.org/10.1161/CIRCULATIONAHA.110.004622
• ²
  Koester MC. A review of sudden cardiac death in young athletes and strategies for preparticipation cardiovascular screening. J Athl Train. 2001;36(2):197-204. PMID: 12937463
• ³
  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(13):1593-601. https://doi.org/10.1001/jama.296.13.1593
  » https://doi.org/10.1001/jama.296.13.1593
• ⁴
  Pelliccia A, Fagard R, Bjørnstad HH, Anastassakis A, Arbustini E, Assanelli D, et al. Recommendations for competitive sports participation in athletes with cardiovascular disease: a consensus document from the Study Group of Sports 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(14):1422-45. https://doi.org/10.1093/eurheartj/ehi325
  » https://doi.org/10.1093/eurheartj/ehi325
• ⁵
  Corrado D, Pelliccia A, Heidbuchel H, Sharma S, Link M, Basso C, et al. Recommendations for interpretation of 12-lead electrocardiogram in the athlete. Eur Heart J. 2010;31(2):243-59. https://doi.org/10.1093/eurheartj/ehp473
  » https://doi.org/10.1093/eurheartj/ehp473
• ⁶
  Drezner JA, Ackerman MJ, Anderson J, Ashley E, Asplund CA, Baggish AL, et al. Electrocardiographic interpretation in athletes: the ‘Seattle criteria’. Br J Sports Med. 2013;47(3):122-4. https://doi.org/10.1136/bjsports-2012-092067
  » https://doi.org/10.1136/bjsports-2012-092067
• ⁷
  Drezner JA, Sharma S, Baggish A, Papadakis M, Wilson MG, Prutkin JM, et al. International criteria for electrocardiographic interpretation in athletes: consensus statement. Br J Sports Med. 2017;51(9):704-31. https://doi.org/10.1136/bjsports-2016-097331
  » https://doi.org/10.1136/bjsports-2016-097331
• ⁸
  BTL Corporate. Prevention of sudden cardiac death [Internet]. [cited on 2021 Oct 20]. Available from: https://www.btlnet.com/products-cardiology-sudden-death-screening
  » https://www.btlnet.com/products-cardiology-sudden-death-screening
• ⁹
  Slaby K, Prochazka M, Kubus P, Mohyla F, Stranik A, Radvansky J, et al. 49th Annual Meeting of the Association for European Paediatric and Congenital Cardiology, AEPC with joint sessions with the Japanese Society of Pediatric Cardiology and Cardiac Surgery, Asia-Pacific Pediatric Cardiology Society, European Association for Cardio-Thoracic Surgery and Canadian Pediatric Cardiology Association, Prague, Czech Republic, 20–23 May 2015. Cardiol Young [Internet]. 2015 [cited on 2021 Oct 20];25(S1):S1-180. Available from: https://www.cambridge.org/core/journals/cardiology-in-the-young/article/49th-annual-meeting-of-the-association-for-european-paediatric-and-congenital-cardiology-aepc-with-joint-sessions-with-the-japanese-society-of-pediatric-cardiology-and-cardiac-surgery-asiapacific-pediatric-cardiology-society-european-association-for-cardiothoracic-surgery-and-canadian-pediatric-cardiology-association-prague-czech-republic-2023-may-2015/0E364FEDA7B81C18532E01044C342D6B
  » https://www.cambridge.org/core/journals/cardiology-in-the-young/article/49th-annual-meeting-of-the-association-for-european-paediatric-and-congenital-cardiology-aepc-with-joint-sessions-with-the-japanese-society-of-pediatric-cardiology-and-cardiac-surgery-asiapacific-pediatric-cardiology-society-european-association-for-cardiothoracic-surgery-and-canadian-pediatric-cardiology-association-prague-czech-republic-2023-may-2015/0E364FEDA7B81C18532E01044C342D6B
• ¹⁰
  Hyde N, Prutkin JM, Drezner JA. Electrocardiogram interpretation in NCAA athletes: comparison of the ‘Seattle’ and ‘International’ criteria. J Electrocardiol. 2019;56:81-4. https://doi.org/10.1016/j.jelectrocard.2019.07.001
  » https://doi.org/10.1016/j.jelectrocard.2019.07.001
• ¹¹
  Gati S, Sheikh N, Ghani S, Zaidi A, Wilson M, Raju H, et al. Should axis deviation or atrial enlargement be categorised as abnormal in young athletes? The athlete's electrocardiogram: time for re-appraisal of markers of pathology. Eur Heart J. 2013;34(47):3641-8. https://doi.org/10.1093/eurheartj/eht390
  » https://doi.org/10.1093/eurheartj/eht390
• ¹²
  Priori SG, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M, et al. Risk stratification in the long-QT syndrome. N Engl J Med. 2003;348(19):1866-74. https://doi.org/10.1056/NEJMoa022147
  » https://doi.org/10.1056/NEJMoa022147
• ¹³
  Georgijević L, Andrić L. Electrocardiography in pre-participation screening and current guidelines for participation in competitive sports. Srp Arh Celok Lek. 2016;144(1-2):104-10. https://doi.org/10.2298/sarh1602104g
  » https://doi.org/10.2298/sarh1602104g
• ¹⁴
  Viskin S, Rosovski U, Sands AJ, Chen E, Kistler PM, Kalman JM, et al. Inaccurate electrocardiographic interpretation of long QT: the majority of physicians cannot recognize a long QT when they see one. Heart Rhythm. 2005;2(6):569-74. https://doi.org/10.1016/j.hrthm.2005.02.011
  » https://doi.org/10.1016/j.hrthm.2005.02.011
• ¹⁵
  Chen AS, Bent RE, Wheeler M, Knowles JW, Haddad F, Froelicher V, et al. Large Q and S waves in lead III on the electrocardiogram distinguish patients with hypertrophic cardiomyopathy from athletes. Heart. 2018;104(22):1871-7. https://doi.org/10.1136/heartjnl-2017-312647
  » https://doi.org/10.1136/heartjnl-2017-312647
• ¹⁶
  Zorzi A, Calore C, Vio R, Pelliccia A, Corrado D. Accuracy of the ECG for differential diagnosis between hypertrophic cardiomyopathy and athlete's heart: comparison between the European Society of Cardiology (2010) and International (2017) criteria. Br J Sports Med. 2018;52(10):667-73. https://doi.org/10.1136/bjsports-2016-097438
  » https://doi.org/10.1136/bjsports-2016-097438
• ¹⁷
  Ng CT, Chee TS, Ling LF, Lee YP, Ching CK, Chua TS, et al. Prevalence of hypertrophic cardiomyopathy on an electrocardiogram-based pre-participation screening programme in a young male South-East Asian population: results from the Singapore Armed Forces Electrocardiogram and Echocardiogram screening protocol. Europace. 2011;13(6):883-8. https://doi.org/10.1093/europace/eur051
  » https://doi.org/10.1093/europace/eur051
• ¹⁸
  Asif IM, Harmon KG. Incidence and etiology of sudden cardiac death: new updates for athletic departments. Sports Health. 2017;9(3):268-79. https://doi.org/10.1177/1941738117694153
  » https://doi.org/10.1177/1941738117694153
• ¹⁹
  Ramos R, Albert X, Sala J, Garcia-Gil M, Elosua R, Marrugat J, et al. Prevalence and incidence of Q-wave unrecognized myocardial infarction in general population: diagnostic value of the electrocardiogram. The REGICOR study. Int J Cardiol. 2016;225:300-5. https://doi.org/10.1016/j.ijcard.2016.10.005
  » https://doi.org/10.1016/j.ijcard.2016.10.005
• ²⁰
  Smulyan H. The computerized ECG: friend and foe. Am J Med. 2019;132(2):153-60. https://doi.org/10.1016/j.amjmed.2018.08.025
  » https://doi.org/10.1016/j.amjmed.2018.08.025

Publication Dates

• Publication in this collection
  27 Oct 2023
• Date of issue
  2023

History

• Received
  10 May 2023
• Accepted
  22 Aug 2023