- Email: [email protected]
Are the QRS duration and ST depression cut-points from the Seattle criteria too conservative?
Available online at www.sciencedirect.com
ScienceDirect Journal of Electrocardiology 48 (2015) 395 – 398 www.jecgonline.com
Are the QRS duration and ST depression cut-points from the Seattle criteria too conservative?☆ Tim Dunn, MD, a,⁎ Ramy Abdelfattah, MD, a Sonya Aggarwal, BA, a David Pickham, RN, PhD, a David Hadley, PhD, b Victor Froelicher, MD a a
Stanford University Medical School, Stanford University, Stanford, CA b Cardea Ltd, Seattle, WA
Abstract
Background: Screening athletes with ECGs is aimed at identifying "at-risk" individuals who may have a cardiac condition predisposing them to sudden cardiac death. The Seattle criteria highlight QRS duration greater than 140 ms and ST segment depression in two or more leads greater than 50 μV as two abnormal ECG patterns associated with sudden cardiac death. Methods: High school, college, and professional athletes underwent 12 lead ECGs as part of routine pre-participation physicals. Prevalence of prolonged QRS duration was measured using cut-points of 120, 125, 130, and 140 ms. ST segment depression was measured in all leads except leads III, aVR, and V1 with cut-points of 25 μV and 50 μV. Results: Between June 2010 and November 2013, 1595 participants including 297 (167 male, mean age 16.2) high school athletes, 1016 (541 male, mean age 18.8) college athletes, and 282 (mean age 26.6) male professional athletes underwent screening with an ECG. Only 3 athletes (0.2%) had a QRS duration greater than 125 ms. ST segment depression in two or more leads greater than 50 μV was uncommon (0.8%), while the prevalence of ST segment depression in two or more leads increased to 4.5% with a cut-point of 25 μV. Conclusion: Changing the QRS duration cut-point to 125 ms would increase the sensitivity of the screening ECG, without a significant increase in false-positives. However, changing the ST segment depression cut-point to 25 μV would lead to a significant increase in false-positives and would therefore not be justified. © 2015 Elsevier Inc. All rights reserved.
Keywords:
Seattle criteria; QRS duration; ST depression
Introduction Preparticipation examination of athletes prior to their participation in competition is aimed at preventing sudden cardiac death. The addition of the screening electrocardiogram (ECG) to the preparticipation examination has been controversial given concern for a high rate of false-positives which lead to unnecessary testing and consultation before an athlete returns to competition. In February 2012, an international group of experts in sports cardiology and sports medicine convened in Seattle, WA, to create standardized criteria for interpretation of ECGs in athletes [1]. The objective of the Seattle criteria was ☆
Disclosures: Victor Froelicher and David Hadley, Co-owners, Cardea Associates (Seattle, Wash). ⁎ Corresponding author at: 300 Pasteur Drive #H2155, Stanford, CA 94305-5233. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jelectrocard.2015.02.009 0022-0736/© 2015 Elsevier Inc. All rights reserved.
to reduce false-positives by distinguishing normal and physiologic ECG patterns commonly found in athletes from abnormal ECG findings representative of conditions that predispose to sudden cardiac death [1]. QRS duration greater than 140 ms and ST segment depression greater than 50 μV in two or more leads are two ECG abnormalities identified by the Seattle criteria that suggest an underlying cardiomyopathy [1]. This paper analyzes different QRS duration and ST segment depression cut-points with the goal of maximizing the sensitivity of the screening ECG without a large increase in false-positives.
Methods Study population Electrocardiograms were obtained from 1595 athletes as part of their preparticipation examination. Testing was
396
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398
performed between June 2010 and November 2013. Consent was obtained from each participant via a consent form approved by the Stanford Investigation Review Board. All athletes received a complete health assessment, electrocardiogram, and follow-up by an expert cardiologist specializing in sports medicine. All of the collegiate and professional athletes had an echocardiogram either as a scanning study or complete study when any abnormality was encountered during the preparticipation examination. No athlete was found to have a cardiovascular condition precluding their participation from competition. After an average follow-up of 2 years, no athlete has experienced an adverse cardiac event. ECG recording and analysis High resolution 16-s 12-lead ECGs were recorded using the CardeaScreen electrocardiograph device (Seattle, WA). Data were recorded at 1 kHz and band-pass filtered (0.05 to 150 Hz). Signal processing and automated measurements and interpretations were completed in near real-time by the CardeaScreen recording device. Computer measurements were made on a median mathematical construct of three orthogonal leads (V5, aVF, and V2) overlaid using autocorrelation signal averaging after removal of noise and ectopic beats. The QRS onsets and offsets were confirmed by visual over-reading and corrected when in error (less than 1% of ECGs). ST measurements and QRS offset were determined at QRS termination which was considered to occur after J waves and slurs if they were present. The ST level was measured irrespective of the slope of the ST segment per the Seattle criteria [1]. ST depressions in this paper were calculated based on computerized measurements that are not subject to the bias of visual measurements which include rounding to the closest amplitude or timing line on the ECG paper. The measurements by the CardeaScreen device have been validated by comparison with another FDA approved commercial device [2]. All measurements and automated ECG interpretations were manually reviewed for accuracy. To determine the ECG values within this cohort, automated measures were imported into a central database where descriptive statistics were completed. All analysis was performed using NCSS 9 software (Kaysville, UT).
Results Table 1 presents the demographics of the study population. A total of 1595 athletes participated in this study. By level of competition, 297 were from high school, 1016 were from college, and 282 were professional athletes. Average age was 19.6 years and 990 (62%) were male. The most common sports were football, crew/rowing, and water sports such as swimming and water polo. Table 2 presents the number and prevalence of athletes with an abnormal QRS duration based on the cut-points 120 ms, 125 ms, 130 ms, and 140 ms. Overall, 21 athletes (1.3%) had a prolonged QRS duration greater than 120 ms while only 3 athletes (0.2%) in this population had a QRS
Table 1 Demographics. N (%) or mean ± SD Total population Level of competition High school College Professional Male African American Age (years) BMI (kg/m2) Heart rate (bpm) Sport Football Crew/rowing Water sports Field hockey, soccer, lacrosse Track and Field Basketball Other
1595 297 (19%) 1016 (64%) 282 (18%) 990 (62%) 306 (19%) 19.6 ± 3.8 24.4 ± 5.2 63.1 ± 11.8 436 (27%) 193 (12%) 154 (10%) 133 (8%) 143 (9%) 96 (6%) 440 (28%)
duration greater than 125 ms. Only one athlete had a QRS duration greater than 130 ms. Tables 3 and 4 present the number and prevalence of athletes with ST segment depression greater than 25 μV and 50 μV, respectively. Each table includes the data for all individual leads except aVR, III, and V1. Leads aVR, III, and V1 were not included in this analysis as these leads are generally excluded when analyzing repolarization phenomena [1]. Additionally, each table includes the number and prevalence of athletes with ST segment depression in two or more lateral leads (I, aVL, V4–V6), inferior leads (II, aVF), and anterior leads (V2–V4), as well as ST segment depression in two or more leads with all leads considered (I, II, aVF, aVL, V2–V6). ST segment depression greater than 25 μV occurred most often in leads aVF (5.3%), aVL (3.8%), and II (3.6%). In all other leads, ST segment depression greater than 25 μV had a prevalence between 0.5% and 1.5%. Conversely, ST segment depression greater than 50 μV was relatively uncommon, with all leads having a prevalence less than 1%. As a group, ST segment depression greater than 25 μV in two or more leads occurred more often in the lateral (1.6%) and inferior leads (2.9%), compared to the anterior leads (0.3%). ST segment depression greater than 50 μV in two or more leads had a similar prevalence between the lateral (0.1%), inferior (0.6%), and anterior (0.1%) leads.
Table 2 Number and prevalence of abnormal QRS duration using different cut-points. N 120 N 125 N 130 N 140
ms ms ms ms
N
%
21 3 1 1
1.3% 0.2% 0.1% 0.1%
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398 Table 3 Number and prevalence of ST segment depression greater than 25 μV.
Lead I Lead II Lead aVL Lead aVF Lead V2 Lead V3 Lead V4 Lead V5 Lead V6 ST depression in two or more lateral leads (I, aVL, V4–V6) ST depression in two or more inferior leads (II, aVF) ST depression in two or more anterior leads (V2–V4) ST depression in any two or more leads (I, II, aVF, aVL, V2–V6)
N
%
14 57 61 84 15 18 8 19 24 26
0.9% 3.6% 3.8% 5.3% 0.9% 1.1% 0.5% 1.2% 1.5% 1.6%
47
2.9%
4
0.3%
71
4.5%
397
Table 4 Number and prevalence of ST segment depression greater than 50 μV.
Lead I Lead II Lead aVL Lead aVF Lead V2 Lead V3 Lead V4 Lead V5 Lead V6 ST depression in two or more lateral leads (I, aVL, V4–V6) ST depression in two or more inferior leads (II, aVF) ST depression in two or more anterior leads (V2–V4) ST depression in any two or more leads (I, II, aVF, aVL, V2–V6)
N
%
2 10 1 15 6 9 3 1 3 2
0.1% 0.6% 0.1% 0.9% 0.4% 0.6% 0.2% 0.1% 0.2% 0.1%
10
0.6%
1
0.1%
12
0.8%
25 μV equals 0.25 mm on typical ECG paper with amplitude set at standard display of 10 mm = 1 mV.
50 μV equals 0.5 mm on typical ECG paper with amplitude set at standard display of 10 mm = 1 mV.
With all leads considered, ST segment depression greater than 25 μV in two or more leads occurred in 71 athletes (4.5%). Conversely, ST segment depression greater than 50 μV in two or more leads was much less common, occurring in only 12 athletes (0.8%).
cardiomyopathy [5]. Their results found that a QRS duration greater than 120 ms was independently associated with increased risk of cardiovascular death. The presence of QRS duration and ST segment depression on a resting ECG has also been associated with increased mortality in the general medical population. In 2014, Perino et al. analyzed computerized ST segment depressions in the resting ECG of 43,661 veterans [6]. Their analysis found that the relative risk of cardiovascular death for ST segment depression N 50 μV ranged from 3.2 to 5.5 in the lateral leads. Conversely, ST segment depression isolated to the anterior or inferior leads, without concomitant lateral depressions, was found to be a poor predictor of cardiovascular death. In 2006, Desai et al. analyzed computerized QRS duration in the resting ECG of 46,933 veterans [7]. Their analysis found that QRS duration was a strong independent predictor of cardiovascular mortality, with every 10 ms increase in QRS duration associated with an 18% increase in cardiovascular risk. Given their association with hypertrophic cardiomyopathy and sudden cardiac death, QRS prolongation and ST segment depression are two ECG patterns that are helpful in identifying “at-risk” athletes. As mentioned earlier, the Seattle criteria define QRS prolongation as greater than 140 ms and ST segment depression as greater than 50 μV in any two leads. The question becomes should the cut-points for QRS duration and ST segment in the Seattle criteria be changed to be more sensitive given their association with sudden cardiac death. According to our results, reducing the QRS duration cut-point from 140 ms to 125 ms should increase the sensitivity of the screening ECG without a large increase in false-positives, as only 0.2% of our study population had a QRS duration greater than 125 ms. While changing the ST segment depression cut-point from 50 μV to 25 μV would likely increase the sensitivity of the screening ECG, it would likely be accompanied by a significant increase in false-positives as 4.5% of our study
Discussion Screening athletes with ECGs is aimed at identifying “at-risk” individuals who may have a cardiac condition predisposing them to sudden cardiac death. The current recommendations for ECG screening criteria are largely based on the “normal” prevalence of ECG patterns in athletes. Because of the relatively rare and infrequent nature of adverse cardiac events in athletes, few studies have had the opportunity to evaluate the connection between particular ECG patterns and actual occurrences of sudden cardiac death. In 2004, Wisten et al. investigated the frequency of ECG abnormalities in young adults who suffered a sudden cardiac death in Sweden between 1992 and 1999 [3]. They found 181 cases of sudden cardiac death, of which 66 people had an ECG at some point prior to their death. The most common ECG patterns were T wave abnormalities (35%), ST segment changes (32%), and conduction defects (20%). This study concluded that pathological ECG patterns, including QRS prolongation and ST segment changes, were common in young victims of sudden cardiac death. Additional studies have demonstrated the association between hypertrophic cardiomyopathy and the ECG patterns of prolonged QRS duration and ST segment depression. In 2009, Haghjoo et al. studied the relationship between ECG findings and prognosis in patients with hypertrophic cardiomyopathy [4]. Their results found that ST segment depression in the high lateral leads was a predictor of sudden cardiac death in patients with hypertrophic cardiomyopathy. Similarly, Bongioanni et al. studied the relationship between QRS duration and prognosis in patients with hypertrophic
398
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398
population had two or more leads with ST segment depression greater than 25 μV. As such, changing the ST segment depression cut-point to 25 μV for all leads would not be justified given the high rate of false-positives. A distinction should be made with regards to ST segment depression in the lateral leads as the above studies have shown a higher association with cardiovascular death in the lateral leads as opposed to the anterior and inferior leads. The importance of ST segment depression in the lateral leads may be explained in that ischemia, myocardial fibrosis, and wall stress are usually a subendocardial phenomenon. They result in an ST vector along the long axis of the left ventricle which is parallel to the lateral leads and perpendicular to the anterior and posterior leads. Given the association between ST segment depression in the lateral leads and cardiovascular death, one could consider adjusting the ST segment depression cut-point to 25 μV for only the lateral leads (I, aVL, V4-V6). Changing the ST segment depression cut-point in the lateral leads to 25 μV would likely increase sensitivity of the screening ECG, with only a modest increase the rate of athletes who screen positive as only 1.6% of our patient population had two or more ST segment depressions greater than 25 μV in the lateral leads. While this could be reduced by considering slope, the added complexity may create reproducibility issues for routine visual interpretation. Limitations Our study is a preliminary analysis focused on the possibilities of improving the screening ECG. Conclusions cannot be made regarding the exact sensitivity and specificity of the screening ECG due to the limited number of participants and lack of adverse outcomes. The study population was limited to our local area and may not be generalizable to other regions.
Conclusions The Seattle criteria consider QRS duration greater than 140 ms and ST segment depression in two or more leads greater than 50 μV as two ECG patterns requiring further evaluation prior to participation in sports. We propose changing the QRS duration cut-point to 125 ms as such a change would likely increase the sensitivity of the screening and would not be accompanied by a significant increase in false-positives. Conversely, changing the ST segment depression cut-point to 25 μV would lead to a significant increase in false-positives, and would therefore not be justified. References [1] Drezner J, Ackerman M, Anderson J, Ashley E, Asplund C, Baggish A, et al. Electrocardiographic interpretation in athletes: the ‘Seattle Criteria’. Br J Sports Med 2013;47:122–4. [2] Muramoto D, Yong D, Singh N, Aggarwal S, Perez M, Ashley E, et al. Patterns and prognosis of all components of the J-wave pattern in multiethnic athletes and ambulatory patients. Am Heart J 2014;167(2):259–66. [3] Wisten A, Andersson S, Forsberg H, Krantz P, Messner T. Sudden cardiac death in the young in Sweden: electrocardiogram in relation to forensic diagnosis. J Intern Med 2004;255:213–20. [4] Haghjoo M, Mohammadzadeh S, Taherpour M, Faghfurian B, Fazelifar A, Alizadeh A, et al. ST-segment depression as a risk factor in hypertrophic cardiomyopathy. Europace 2009;11(5):643–9. [5] Bongioanni S, Bianchi F, Migliardi A, Gnavi R, Pron P, Casetta M, et al. Relation of QRS duration to mortality in a community-based cohort with hypertrophic cardiomyopathy. Am J Cardiol 2007;100:503–6. [6] Perino A, Singh N, Aggarwal S, Froelicher V. The long-term prognostic value of the ST depression criteria for ischemia recommended in the universal definition of myocardial infarction in 43,661 veterans. Int J Cardiol 2014;173:494–8. [7] Desai A, Yaw T, Yamazaki T, Kayhka A, Chun S, Froelicher V. Prognostic significance of quantitative QRS duration. Am J Med 2006;119:600–6.
ScienceDirect Journal of Electrocardiology 48 (2015) 395 – 398 www.jecgonline.com
Are the QRS duration and ST depression cut-points from the Seattle criteria too conservative?☆ Tim Dunn, MD, a,⁎ Ramy Abdelfattah, MD, a Sonya Aggarwal, BA, a David Pickham, RN, PhD, a David Hadley, PhD, b Victor Froelicher, MD a a
Stanford University Medical School, Stanford University, Stanford, CA b Cardea Ltd, Seattle, WA
Abstract
Background: Screening athletes with ECGs is aimed at identifying "at-risk" individuals who may have a cardiac condition predisposing them to sudden cardiac death. The Seattle criteria highlight QRS duration greater than 140 ms and ST segment depression in two or more leads greater than 50 μV as two abnormal ECG patterns associated with sudden cardiac death. Methods: High school, college, and professional athletes underwent 12 lead ECGs as part of routine pre-participation physicals. Prevalence of prolonged QRS duration was measured using cut-points of 120, 125, 130, and 140 ms. ST segment depression was measured in all leads except leads III, aVR, and V1 with cut-points of 25 μV and 50 μV. Results: Between June 2010 and November 2013, 1595 participants including 297 (167 male, mean age 16.2) high school athletes, 1016 (541 male, mean age 18.8) college athletes, and 282 (mean age 26.6) male professional athletes underwent screening with an ECG. Only 3 athletes (0.2%) had a QRS duration greater than 125 ms. ST segment depression in two or more leads greater than 50 μV was uncommon (0.8%), while the prevalence of ST segment depression in two or more leads increased to 4.5% with a cut-point of 25 μV. Conclusion: Changing the QRS duration cut-point to 125 ms would increase the sensitivity of the screening ECG, without a significant increase in false-positives. However, changing the ST segment depression cut-point to 25 μV would lead to a significant increase in false-positives and would therefore not be justified. © 2015 Elsevier Inc. All rights reserved.
Keywords:
Seattle criteria; QRS duration; ST depression
Introduction Preparticipation examination of athletes prior to their participation in competition is aimed at preventing sudden cardiac death. The addition of the screening electrocardiogram (ECG) to the preparticipation examination has been controversial given concern for a high rate of false-positives which lead to unnecessary testing and consultation before an athlete returns to competition. In February 2012, an international group of experts in sports cardiology and sports medicine convened in Seattle, WA, to create standardized criteria for interpretation of ECGs in athletes [1]. The objective of the Seattle criteria was ☆
Disclosures: Victor Froelicher and David Hadley, Co-owners, Cardea Associates (Seattle, Wash). ⁎ Corresponding author at: 300 Pasteur Drive #H2155, Stanford, CA 94305-5233. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jelectrocard.2015.02.009 0022-0736/© 2015 Elsevier Inc. All rights reserved.
to reduce false-positives by distinguishing normal and physiologic ECG patterns commonly found in athletes from abnormal ECG findings representative of conditions that predispose to sudden cardiac death [1]. QRS duration greater than 140 ms and ST segment depression greater than 50 μV in two or more leads are two ECG abnormalities identified by the Seattle criteria that suggest an underlying cardiomyopathy [1]. This paper analyzes different QRS duration and ST segment depression cut-points with the goal of maximizing the sensitivity of the screening ECG without a large increase in false-positives.
Methods Study population Electrocardiograms were obtained from 1595 athletes as part of their preparticipation examination. Testing was
396
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398
performed between June 2010 and November 2013. Consent was obtained from each participant via a consent form approved by the Stanford Investigation Review Board. All athletes received a complete health assessment, electrocardiogram, and follow-up by an expert cardiologist specializing in sports medicine. All of the collegiate and professional athletes had an echocardiogram either as a scanning study or complete study when any abnormality was encountered during the preparticipation examination. No athlete was found to have a cardiovascular condition precluding their participation from competition. After an average follow-up of 2 years, no athlete has experienced an adverse cardiac event. ECG recording and analysis High resolution 16-s 12-lead ECGs were recorded using the CardeaScreen electrocardiograph device (Seattle, WA). Data were recorded at 1 kHz and band-pass filtered (0.05 to 150 Hz). Signal processing and automated measurements and interpretations were completed in near real-time by the CardeaScreen recording device. Computer measurements were made on a median mathematical construct of three orthogonal leads (V5, aVF, and V2) overlaid using autocorrelation signal averaging after removal of noise and ectopic beats. The QRS onsets and offsets were confirmed by visual over-reading and corrected when in error (less than 1% of ECGs). ST measurements and QRS offset were determined at QRS termination which was considered to occur after J waves and slurs if they were present. The ST level was measured irrespective of the slope of the ST segment per the Seattle criteria [1]. ST depressions in this paper were calculated based on computerized measurements that are not subject to the bias of visual measurements which include rounding to the closest amplitude or timing line on the ECG paper. The measurements by the CardeaScreen device have been validated by comparison with another FDA approved commercial device [2]. All measurements and automated ECG interpretations were manually reviewed for accuracy. To determine the ECG values within this cohort, automated measures were imported into a central database where descriptive statistics were completed. All analysis was performed using NCSS 9 software (Kaysville, UT).
Results Table 1 presents the demographics of the study population. A total of 1595 athletes participated in this study. By level of competition, 297 were from high school, 1016 were from college, and 282 were professional athletes. Average age was 19.6 years and 990 (62%) were male. The most common sports were football, crew/rowing, and water sports such as swimming and water polo. Table 2 presents the number and prevalence of athletes with an abnormal QRS duration based on the cut-points 120 ms, 125 ms, 130 ms, and 140 ms. Overall, 21 athletes (1.3%) had a prolonged QRS duration greater than 120 ms while only 3 athletes (0.2%) in this population had a QRS
Table 1 Demographics. N (%) or mean ± SD Total population Level of competition High school College Professional Male African American Age (years) BMI (kg/m2) Heart rate (bpm) Sport Football Crew/rowing Water sports Field hockey, soccer, lacrosse Track and Field Basketball Other
1595 297 (19%) 1016 (64%) 282 (18%) 990 (62%) 306 (19%) 19.6 ± 3.8 24.4 ± 5.2 63.1 ± 11.8 436 (27%) 193 (12%) 154 (10%) 133 (8%) 143 (9%) 96 (6%) 440 (28%)
duration greater than 125 ms. Only one athlete had a QRS duration greater than 130 ms. Tables 3 and 4 present the number and prevalence of athletes with ST segment depression greater than 25 μV and 50 μV, respectively. Each table includes the data for all individual leads except aVR, III, and V1. Leads aVR, III, and V1 were not included in this analysis as these leads are generally excluded when analyzing repolarization phenomena [1]. Additionally, each table includes the number and prevalence of athletes with ST segment depression in two or more lateral leads (I, aVL, V4–V6), inferior leads (II, aVF), and anterior leads (V2–V4), as well as ST segment depression in two or more leads with all leads considered (I, II, aVF, aVL, V2–V6). ST segment depression greater than 25 μV occurred most often in leads aVF (5.3%), aVL (3.8%), and II (3.6%). In all other leads, ST segment depression greater than 25 μV had a prevalence between 0.5% and 1.5%. Conversely, ST segment depression greater than 50 μV was relatively uncommon, with all leads having a prevalence less than 1%. As a group, ST segment depression greater than 25 μV in two or more leads occurred more often in the lateral (1.6%) and inferior leads (2.9%), compared to the anterior leads (0.3%). ST segment depression greater than 50 μV in two or more leads had a similar prevalence between the lateral (0.1%), inferior (0.6%), and anterior (0.1%) leads.
Table 2 Number and prevalence of abnormal QRS duration using different cut-points. N 120 N 125 N 130 N 140
ms ms ms ms
N
%
21 3 1 1
1.3% 0.2% 0.1% 0.1%
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398 Table 3 Number and prevalence of ST segment depression greater than 25 μV.
Lead I Lead II Lead aVL Lead aVF Lead V2 Lead V3 Lead V4 Lead V5 Lead V6 ST depression in two or more lateral leads (I, aVL, V4–V6) ST depression in two or more inferior leads (II, aVF) ST depression in two or more anterior leads (V2–V4) ST depression in any two or more leads (I, II, aVF, aVL, V2–V6)
N
%
14 57 61 84 15 18 8 19 24 26
0.9% 3.6% 3.8% 5.3% 0.9% 1.1% 0.5% 1.2% 1.5% 1.6%
47
2.9%
4
0.3%
71
4.5%
397
Table 4 Number and prevalence of ST segment depression greater than 50 μV.
Lead I Lead II Lead aVL Lead aVF Lead V2 Lead V3 Lead V4 Lead V5 Lead V6 ST depression in two or more lateral leads (I, aVL, V4–V6) ST depression in two or more inferior leads (II, aVF) ST depression in two or more anterior leads (V2–V4) ST depression in any two or more leads (I, II, aVF, aVL, V2–V6)
N
%
2 10 1 15 6 9 3 1 3 2
0.1% 0.6% 0.1% 0.9% 0.4% 0.6% 0.2% 0.1% 0.2% 0.1%
10
0.6%
1
0.1%
12
0.8%
25 μV equals 0.25 mm on typical ECG paper with amplitude set at standard display of 10 mm = 1 mV.
50 μV equals 0.5 mm on typical ECG paper with amplitude set at standard display of 10 mm = 1 mV.
With all leads considered, ST segment depression greater than 25 μV in two or more leads occurred in 71 athletes (4.5%). Conversely, ST segment depression greater than 50 μV in two or more leads was much less common, occurring in only 12 athletes (0.8%).
cardiomyopathy [5]. Their results found that a QRS duration greater than 120 ms was independently associated with increased risk of cardiovascular death. The presence of QRS duration and ST segment depression on a resting ECG has also been associated with increased mortality in the general medical population. In 2014, Perino et al. analyzed computerized ST segment depressions in the resting ECG of 43,661 veterans [6]. Their analysis found that the relative risk of cardiovascular death for ST segment depression N 50 μV ranged from 3.2 to 5.5 in the lateral leads. Conversely, ST segment depression isolated to the anterior or inferior leads, without concomitant lateral depressions, was found to be a poor predictor of cardiovascular death. In 2006, Desai et al. analyzed computerized QRS duration in the resting ECG of 46,933 veterans [7]. Their analysis found that QRS duration was a strong independent predictor of cardiovascular mortality, with every 10 ms increase in QRS duration associated with an 18% increase in cardiovascular risk. Given their association with hypertrophic cardiomyopathy and sudden cardiac death, QRS prolongation and ST segment depression are two ECG patterns that are helpful in identifying “at-risk” athletes. As mentioned earlier, the Seattle criteria define QRS prolongation as greater than 140 ms and ST segment depression as greater than 50 μV in any two leads. The question becomes should the cut-points for QRS duration and ST segment in the Seattle criteria be changed to be more sensitive given their association with sudden cardiac death. According to our results, reducing the QRS duration cut-point from 140 ms to 125 ms should increase the sensitivity of the screening ECG without a large increase in false-positives, as only 0.2% of our study population had a QRS duration greater than 125 ms. While changing the ST segment depression cut-point from 50 μV to 25 μV would likely increase the sensitivity of the screening ECG, it would likely be accompanied by a significant increase in false-positives as 4.5% of our study
Discussion Screening athletes with ECGs is aimed at identifying “at-risk” individuals who may have a cardiac condition predisposing them to sudden cardiac death. The current recommendations for ECG screening criteria are largely based on the “normal” prevalence of ECG patterns in athletes. Because of the relatively rare and infrequent nature of adverse cardiac events in athletes, few studies have had the opportunity to evaluate the connection between particular ECG patterns and actual occurrences of sudden cardiac death. In 2004, Wisten et al. investigated the frequency of ECG abnormalities in young adults who suffered a sudden cardiac death in Sweden between 1992 and 1999 [3]. They found 181 cases of sudden cardiac death, of which 66 people had an ECG at some point prior to their death. The most common ECG patterns were T wave abnormalities (35%), ST segment changes (32%), and conduction defects (20%). This study concluded that pathological ECG patterns, including QRS prolongation and ST segment changes, were common in young victims of sudden cardiac death. Additional studies have demonstrated the association between hypertrophic cardiomyopathy and the ECG patterns of prolonged QRS duration and ST segment depression. In 2009, Haghjoo et al. studied the relationship between ECG findings and prognosis in patients with hypertrophic cardiomyopathy [4]. Their results found that ST segment depression in the high lateral leads was a predictor of sudden cardiac death in patients with hypertrophic cardiomyopathy. Similarly, Bongioanni et al. studied the relationship between QRS duration and prognosis in patients with hypertrophic
398
T. Dunn et al. / Journal of Electrocardiology 48 (2015) 395–398
population had two or more leads with ST segment depression greater than 25 μV. As such, changing the ST segment depression cut-point to 25 μV for all leads would not be justified given the high rate of false-positives. A distinction should be made with regards to ST segment depression in the lateral leads as the above studies have shown a higher association with cardiovascular death in the lateral leads as opposed to the anterior and inferior leads. The importance of ST segment depression in the lateral leads may be explained in that ischemia, myocardial fibrosis, and wall stress are usually a subendocardial phenomenon. They result in an ST vector along the long axis of the left ventricle which is parallel to the lateral leads and perpendicular to the anterior and posterior leads. Given the association between ST segment depression in the lateral leads and cardiovascular death, one could consider adjusting the ST segment depression cut-point to 25 μV for only the lateral leads (I, aVL, V4-V6). Changing the ST segment depression cut-point in the lateral leads to 25 μV would likely increase sensitivity of the screening ECG, with only a modest increase the rate of athletes who screen positive as only 1.6% of our patient population had two or more ST segment depressions greater than 25 μV in the lateral leads. While this could be reduced by considering slope, the added complexity may create reproducibility issues for routine visual interpretation. Limitations Our study is a preliminary analysis focused on the possibilities of improving the screening ECG. Conclusions cannot be made regarding the exact sensitivity and specificity of the screening ECG due to the limited number of participants and lack of adverse outcomes. The study population was limited to our local area and may not be generalizable to other regions.
Conclusions The Seattle criteria consider QRS duration greater than 140 ms and ST segment depression in two or more leads greater than 50 μV as two ECG patterns requiring further evaluation prior to participation in sports. We propose changing the QRS duration cut-point to 125 ms as such a change would likely increase the sensitivity of the screening and would not be accompanied by a significant increase in false-positives. Conversely, changing the ST segment depression cut-point to 25 μV would lead to a significant increase in false-positives, and would therefore not be justified. References [1] Drezner J, Ackerman M, Anderson J, Ashley E, Asplund C, Baggish A, et al. Electrocardiographic interpretation in athletes: the ‘Seattle Criteria’. Br J Sports Med 2013;47:122–4. [2] Muramoto D, Yong D, Singh N, Aggarwal S, Perez M, Ashley E, et al. Patterns and prognosis of all components of the J-wave pattern in multiethnic athletes and ambulatory patients. Am Heart J 2014;167(2):259–66. [3] Wisten A, Andersson S, Forsberg H, Krantz P, Messner T. Sudden cardiac death in the young in Sweden: electrocardiogram in relation to forensic diagnosis. J Intern Med 2004;255:213–20. [4] Haghjoo M, Mohammadzadeh S, Taherpour M, Faghfurian B, Fazelifar A, Alizadeh A, et al. ST-segment depression as a risk factor in hypertrophic cardiomyopathy. Europace 2009;11(5):643–9. [5] Bongioanni S, Bianchi F, Migliardi A, Gnavi R, Pron P, Casetta M, et al. Relation of QRS duration to mortality in a community-based cohort with hypertrophic cardiomyopathy. Am J Cardiol 2007;100:503–6. [6] Perino A, Singh N, Aggarwal S, Froelicher V. The long-term prognostic value of the ST depression criteria for ischemia recommended in the universal definition of myocardial infarction in 43,661 veterans. Int J Cardiol 2014;173:494–8. [7] Desai A, Yaw T, Yamazaki T, Kayhka A, Chun S, Froelicher V. Prognostic significance of quantitative QRS duration. Am J Med 2006;119:600–6.