CRY Research Fellow, Dr Nabeel Sheikh, had the article ‘Comparison of ECG Criteria for the Detection of Cardiac Abnormalities in Elite Black and White Athletes’ published in Circulation (March 2014). [Abstract here]
CRY Communications Officer, Mair Shepherd, caught up with him to find out more about the research behind this article.
Why did you conduct this research?
A bit of background is required before explaining how this research came about. As many people will know, sudden cardiac death (SCD) is the leading cause of non-traumatic mortality in young (<35 years old) athletes. The vast majority of cases are due to several inherited and congenital cardiac conditions which can be detected by screening. As a consequence, an increasing number of sporting bodies and scientific organisations are now recommending pre-participation screening of athletes prior to clearance to compete.
However, despite what appears by most people to be a laudable aim, pre-participation screening remains a controversial topic. In particular, intense debate exists as to the best way to perform screening. Some authorities, such as the American Heart Association, state that screening should be limited to history and examination; whereas others, such as the European Society of Cardiology (ESC), recommend addition of a 12-lead electrocardiogram (ECG).
One of reservations about ECG screening is false-positive results, which occur because the changes on ECG that can develop as a result of exercise may sometimes overlap with the same changes that occur in conditions responsible for SCD in athletes. False positive results are problematic because they not only create undue worry and anxiety for the athlete, they also add significantly to cost of evaluating an athlete.
To help doctors tell the difference between normal (physiological) and abnormal (pathological) ECG patterns, the ESC produced guidelines in 2005 to help interpretation of an athlete’s ECG. However, although updated in 2010, these guidelines have remained associated with high false positive rates of between 5-20%. Furthermore, the ESC recommendations are derived entirely from data on Caucasian (white) athletes and fail to account for changes which athletes of other ethnicities get – in particular athletes of African/Afro-Caribbean ethnicity (black athletes). Data now clearly shows that black athletes exhibit significantly more ECG changes in response to exercise than white athletes, and that some of these changes may be normal for a black athlete. Current guidelines do not account for these changes, which means that black athletes may be at higher risk of false-positive results and, at worst, false disqualification from sport.
Although publication of the “Seattle Criteria” in 2013 attempted to further help doctors tell the difference between normal and abnormal ECG patterns – and accounted for some of the changes black athletes may develop – these were not all based on research evidence and in our opinion did not go far enough in tackling the problem of false positive results. Our own experience of screening elite athletes through CRY led us to re-evaluate our practice of ECG interpretation in athletes from 2010 onwards, particularly with respect to 6 different ECG patterns that are considered abnormal by current guidelines. Through our research, we found that these 6 ECG patterns do not reflect heart disease in an otherwise healthy athlete; they likely simply reflect the way the heart adapts to intense exercise, and can therefore be considered normal, training related changes.
As a result of the above findings, we came up with a set of “refined” ECG screening criteria. The aim of the current study was to assess the impact of these refined ECG criteria on the false-positive ECG rate in both black and white athletes.
What was the study about?
As mentioned, in the current study we assessed the impact of our refined criteria on the false positive ECG rate in a large number of black (n=1208) and white (n=4297) athletes undergoing pre-participation screening with CRY between 2000 and 2012. The ECGs of all these athletes were re-evaluated using our new refined criteria, the current ESC recommendations and the recent Seattle Criteria, to see how many positive ECG results would be produced by each criteria, that required athletes to undergo further investigations. All 3 ECG criteria were also applied to the ECGs of 103 young, asymptomatic athletes who had already been diagnosed with hypertrophic cardiomyopathy (HCM) to see how many athletes with HCM would be correctly identified by each ECG criteria. It was very important to test each set of criteria in athletes with HCM, as this is the leading cause of death in young athletes worldwide.
What did the results show?
The results showed that the ESC recommendations caused a staggering 40.4% of black athletes to test positive with an abnormal ECG, requiring them to undergo further investigation prior to being given clearance to compete. Importantly, however, almost 1 in 5 white athletes (16.2%) also tested positive on the basis of the ESC recommendations. The Seattle Criteria reduced the number of positive ECGs to 18.4% in black athletes and 7.1% in white athletes. However, the refined criteria further reduced abnormal ECGs to 11.5% in black athletes and just 5.3% in white athletes. Significantly, all 3 criteria identified 98.1% of athletes with HCM.
Overall, the study identified 40 athletes with a cardiac condition. Of these individuals, 25 were diagnosed with only minor problems. The remaining 15 were diagnosed with serious pathology, which was defined as a condition that has been recognized as a cause of exercise-related SCD in young athletes. All 15 cases of serious pathology were identified by a combination of history and 12-lead ECG, with the majority (93.3%) identified on the basis of ECG.
We also calculated the sensitivity and specificity of screening using all 3 criteria. The refined criteria improved specificity in black athletes from 40.3% using the ESC recommendations to 84.2%, and in white athletes from 73.8% using the ESC recommendations to 94.1%. Importantly, sensitivity for detecting all cardiac conditions, including HCM, remained at 70% in black athletes and 60% in white athletes, regardless of the criteria employed. Exclusion of all the minor pathology from our calculations resulted in a dramatic improvement in sensitivity to 100% in both black and white athletes (for all 3 criteria) without a compromise in specificity.
What are the implications of the results?
This crucial work has furthered our understanding of normal versus abnormal ECG patterns in athletes, showing that based on CRY’s novel research, refinement of current ECG screening criteria can have a significant impact in reducing the burden of false positive results during pre-participation screening. Importantly, this occurs without reducing the ECG’s sensitivity for detecting serious cardiac conditions. Indeed, the ECG correctly identified 93.3% of serious cardiac pathology which may otherwise have gone undetected.
Given that the majority of professional sporting organisations utilise ECG-based screening protocols and that no protocol based exclusively on history and examination has demonstrated a reduction in sports-related SCD, surely the next best approach is to strive toward an improvement in screening specificity as we have attempted here? This should be coupled with appropriate training of physicians in correct interpretation of an athlete’s ECG. Future work should focus on further reducing false positive results in black athletes, a significant proportion of whom continue to exhibit positive ECG results.
