- Email: [email protected]
New methodologies for measuring Brugada ECG patterns cannot differentiate the ECG pattern of Brugada syndrome from Brugada phenocopy
Available online at www.sciencedirect.com
ScienceDirect Journal of Electrocardiology xx (2015) xxx – xxx www.jecgonline.com
New methodologies for measuring Brugada ECG patterns cannot differentiate the ECG pattern of Brugada syndrome from Brugada phenocopy☆,☆☆ Byron H. Gottschalk, BMSc, a Javier Garcia-Niebla, RN, b Daniel D. Anselm, MD, c Atul Jaidka, BMSc, a Antoni Bayés De Luna, MD PhD, d Adrian Baranchuk, MD, FACC, FRCPC a,⁎ a
Department of Cardiology, Kingston General Hospital, Queen's University, Kingston, ON, Canada Servicios Sanitarios del Área de Salud de El Hierro, Valle del Golfo Health Center, Canary Island, Spain Libin Cardiovascular Institute of Alberta, Foothills Medical Centre, University of Calgary, Calgary, AB, Canada d Cardiovascular Research Center, CSIC-ICCC, Hospital de la Santa CreuiSant Pau, Barcelona, Spain b
c
Abstract
Background: Brugada phenocopies (BrP) are clinical entities characterized by ECG patterns that are identical to true Brugada syndrome (BrS), but are elicited by various clinical circumstances. A recent study demonstrated that the patterns of BrP and BrS are indistinguishable under the naked eye, thereby validating the concept that the patterns are identical. Objective: The aim of our study was to determine whether recently developed ECG criteria would allow for discrimination between type-2 BrS ECG pattern and type-2 BrP ECG pattern. Methods: Ten ECGs from confirmed BrS (aborted sudden death, transformation into type 1 upon sodium channel blocking test and/or ventricular arrhythmias, positive genetics) cases and 9 ECGs from confirmed BrP were included in the study. Surface 12-lead ECGs were scanned, saved in JPEG format for blind measurement of two values: (i) β-angle; and (ii) the base of the triangle. Cut-off values of ≥ 58° for the β-angle and ≥ 4 mm for the base of the triangle were used to determine the BrS ECG pattern. Results: Mean values for the β-angle in leads V1 and V2 were 66.7 ± 25.5 and 55.4 ± 28.1 for BrS and 54.1 ± 26.5 and 43.1 ± 16.1 for BrP respectively (p = NS). Mean values for the base of the triangle in V1 and V2 were 7.5 ± 3.9 and 5.7 ± 3.9 for BrS and 5.6 ± 3.2 and 4.7 ± 2.7 for BrP respectively (p = NS). The β-angle had a sensitivity of 60%, specificity of 78% (LR + 2.7, LR − 0.5). The base of the triangle had a sensitivity of 80%, specificity of 40% (LR + 1.4, LR − 0.5). Conclusions: New ECG criteria presented relatively low sensitivity and specificity, positive and negative predictive values to discriminate between BrS and BrP ECG patterns, providing further evidence that the two patterns are identical. © 2015 Elsevier Inc. All rights reserved.
Keywords:
Brugada ECG pattern; Brugada phenocopy; Brugada syndrome; ECG criteria; Beta angle; Base of triangle
Introduction Brugada phenocopies (BrP) are clinical entities characterized by ECG patterns presumed to be identical to those of true Brugada syndrome (BrS) but are elicited by various other underlying conditions [1–3]. Differentiation between the two conditions is of the utmost importance as the ☆
Conflict of Interest Disclosure: None. Funding: None. ⁎ Corresponding author at: Heart Rhythm Service, Cardiac Electrophysiology and Pacing, Kingston General Hospital, Queen's University, K7L 2V7 Kingston, ON, Canada. E-mail address: [email protected] ☆☆
http://dx.doi.org/10.1016/j.jelectrocard.2015.12.011 0022-0736/© 2015 Elsevier Inc. All rights reserved.
treatment recommendations of BrP differ from BrS and misdiagnosis may lead to significant morbidity and mortality [4]. While individuals with BrS are at risk for sudden cardiac death (SCD), and may be candidates for an implantable cardioverter-defibrillator (ICD), the clinical implications of BrP remain unknown. Moreover, patients with true congenital BrS should refrain from using drugs with sodium channel blocking properties, as these have been shown to increase the risk of developing potentially fatal arrhythmias [5–7]. Whether the ECG pattern in BrP serves as a warning of increased arrhythmogenicity has yet to be determined. Therefore, current recommendations for the treatment of BrP focus on resolution of the underlying condition [5].
2
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
Table 1 Brugada phenocopy systematic diagnostic criteria (reproduced with permission [5]). Brugada phenocopy diagnostic criteria (first 5 mandatory) i. ii. iii. iv.
ECG pattern has type 1 or type 2 Brugada morphologic characteristics Patient has underlying condition that is identifiable ECG pattern resolves after resolution of the underlying condition There is low clinical pretest probability of true Brugada syndrome determined by lack of symptoms, medical history, and family history v. Negative results on provocative testing with sodium channel blockers such as ajmaline, flecainide, or procainamide (unless clearly not clinically indicated) vi. Provocative testing not mandatory if surgical right ventricular outflow tract manipulation has occurred within last 96 h vii. Results of genetic testing are negative (desirable but not mandatory because the SCN5A mutation is identified in only 20%–30% of probands affected by true Brugada syndrome) Features that suggest true congenital Brugada syndrome i. ECG pattern shows type 1 or type 2 Brugada morphologic characteristics ii. There is a high clinical pretest probability of true congenital Brugada syndrome determined by presence of symptoms, medical history, and family history iii. Positive results on provocative testing with sodium channel blockers such as ajmaline, flecainide, or procainamide; this indicates sodium channel dysfunction consistent with true Brugada syndrome iv. Genetic testing is positive in about 20%–30% of probands ECG, electrocardiographic; SCN5A, sodium channel voltage-gated type V alpha subunit. Adapted from Anselm et al [5].
Diagnosis of BrP relies on a series of electrocardiographic and clinical characteristics that distinguishes the condition from BrS (Table 1). The first mandatory criterion requires an ECG pattern that is identical to the type 1 or type 2 Brugada ECG pattern. However, until recently, this concept had never been validated. A recent study demonstrated that the patterns of BrP and BrS are indistinguishable under the naked eye [8]. The objective of this study was to determine whether measuring the β-angle [9] and the base of the triangle [10] would allow for a distinction between the type 2 BrP and type 2 BrS pattern.
Methods Ten ECGs from confirmed BrS cases (aborted sudden death, transformation into type 1 upon sodium channel blocking test and/or ventricular arrhythmias, positive genetics) all presenting with type 2 ECG Brugada pattern and 9 ECGs from confirmed BrP (again, all presenting type 2 ECG Brugada pattern) were included in the study. Surface 12-lead ECGs were scanned, saved in JPEG format at 150– 300 DPI, and sent to an evaluator (JGN) in randomized order for evaluation, labeled with only a case number. Images were entered into Autodesk AutoCAD® 2013 software for Windows (San Rafael, CA). Once calibrated, two variables were measured: (i) β-angle, defined as the angle between the upslope of the S-wave and the downslope of the r′-wave (Fig. 1A); and (ii) the base of the triangle, defined as the duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV from the high take-off (Fig. 1B). Three complexes were measured in leads V1 and V2 for each case. These values were averaged to give 1 value for each measured variable in leads V1 and V2. Diagnosis was determined completely by measurement, with no clinical history provided. Cut-off values of ≥ 58° for the β-angle and ≥ 4 mm for the base of the triangle were used to determine the Brugada ECG pattern. A case met criteria for a Brugada ECG pattern if either V1 or V2 (or both) exceeded the cut-off values for the variable being measured. Results are described as mean ± SD for continuous variables and percentage (%) for categorical variables. P values b 0.05 were considered statistically significant. Sensitivity, specificity and likelihood ratios for each variable were calculated to determine diagnostic accuracy. Statistical analysis was performed on IBM SPSS software, version 22 for Windows (Armonk, NY, 2014).
Results The average value for the β-angle was 61.3 ± 25.7° in V1 for all cases and 49.3 ± 23.1° in V2 for all cases. The
Fig. 1. Panel A: The β-angle is calculated by measuring the angle between the upslope of the S-wave and the downslope of the r′-wave. Panel B: The base of the triangle is calculated by measuring duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV from the high take-off.
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
average value for the β-angle in V1 was 66.7 ± 25.5° within the BrS group, and 54.1 ± 26.5° within the BrP group (p = 0.85). The average value for the β-angle in V2 was 55.4 ± 28.1° within the BrS group, and 43.1 ± 16.1° within the BrP group (p = 0.29). The average measurement of the base of the triangle was 6.6 ± 3.6 mm in V1 and 5.3 ± 3.3 mm for all cases. The base of the triangle measured on average 7.5 ± 3.9 mm in lead V1 within the BrS group and 5.6 ± 3.2 mm in V1 within the BrP group (p = 0.85). Similarly, the average measurement of the base of the triangle in lead V2 was 5.7 ± 3.9 mm within the BrS group and 4.7 ± 2.7 mm within the BrP group (p = 0.55). With a cut-off of 58°, use of the β-angle to determine the correct diagnosis had a sensitivity of 60%, specificity of 78% with a corresponding positive likelihood ratio of 2.7 and negative likelihood ratio of 0.5. Using a cut-off value of 4 mm, measuring the base of the triangle to determine the correct diagnosis had a sensitivity of 80% and a specificity of 40% with a positive likelihood ratio of 1.4 and a negative likelihood ratio of 0.5 (Table 2).
3
which had a respective sensitivity and specificity of 95.6% and 85%, for the detection of Brugada ECG pattern. Our data suggest that the new methodologies, previously shown to accurately distinguish between a Brugada ECG pattern and patterns of similar morphology (IRBBB or healthy athletes [9,10]), are unable to distinguish a BrS ECG pattern from that of BrP (when both present with type 2 ECG Brugada pattern) (Fig. 2). The difference in the average values for the β-angle and the base of the triangle were not statistically significant between the two groups. Out of the two variables measured, the base of the triangle provided the greatest sensitivity in identifying a Brugada ECG pattern; however the specificity was poor in excluding the BrP ECG patterns. These data suggest a poor ability of the base of the triangle in distinguishing between the ECG patterns of BrS and BrP. While measurements of the β-angle provided an improved specificity when compared to the base of the triangle, the sensitivity was relatively low. Using the β-angle, 11 of the 19 ECG patterns were diagnosed as a non-Brugada ECG pattern, with 4 of the 11 (36%) cases being true BrS. Interestingly; there were no cases that met the criteria for a diagnosis of a Brugada ECG pattern using the β-angle that did not meet the criteria using the base of the triangle. Therefore, when combining both variables in the analysis, there was no improvement of the sensitivity or specificity (Table 3). The critical findings in this study are that neither the base of the triangle nor the β-angle, or any combination of the two variables were able to reliably and accurately distinguish between the ECG patterns of BrP and BrS (Fig. 2). Additionally, the low sensitivity, specificity and weak likelihood ratios provide further evidence that the diagnostic accuracy of BrP and BrS based only on ECG criteria alone are poor. Our data expand on recent evidence that BrP and BrS ECG patterns are identical and indistinguishable [8]. This information reaffirms the notion that systematic diagnostic criteria are indeed required (Table 1). Perhaps the most important component in the reliable distinction between BrS and BrP is the use of a sodium channel blocking agent in a provocative test. Ajmaline has been shown to be more effective than flecainide in the induction or enhancement of the type 1 Brugada ECG pattern [12]. It is therefore recommended that ajmaline be used in these situations whenever possible. Should ajmaline be unavailable, we suggest the use of flecainide as the provocative drug of choice as it has been shown to be more effective than procainamide [13]. This testing is especially important in situations where a clinical condition has been shown to cause BrP as well as unmask BrS such as myocardial ischemia or hyperkalemia [11,14–18].
Discussion In 2011, Chevallier et al. [9] applied newly developed ECG criteria to successfully distinguish between type-2 Brugada ECG pattern and incomplete right bundle branch block (IRBBB) ECG pattern. They measured values of two ECG parameters: (i) the angle between a vertical line and the downslope of the r′-wave (α-angle); and (ii) the angle between the upslope of the S-wave and the downslope of the r′-wave (β-angle) (Fig. 1A) (see Movie 1 in Supplementary Material) [9]. The authors determined that cut-off values of the α- and β-angles at ≥ 50 and ≥ 58° respectively were optimal for the detection of Brugada ECG pattern. Subsequently, Serra et al. [10] validated the use of the β-angle along with three new criteria in the distinction of the ECG of healthy athletes with r′-waves from type-2 Brugada ECG pattern. The three new criteria included: (i) the duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV (5 mm) from the high take-off; (ii) the duration of the base of the triangle at the isoelectric line; and (iii) the ratio of base/height of the triangle formed by the upslope and downslope of the r′-wave at the isoelectric line (Fig. 1B) (see Movie 2 in Supplementary Material) [10]. The easiest parameter to measure from the criteria described by Serra et al. was the duration of the base of the triangle at 5 mm from the high take-off. They determined an optimal cut-off value of ≥ 160 ms (4 mm),
Table 2 Average measurements for BrS compared to BrP. Measurement
BrS V1
BrP V1
P value
BrS V2
BrP V2
P value
Base β-angle
7.5 ± 3.9 66.7 ± 25.5
5.6 ± 3.2 54.1 ± 26.5
0.85 0.85
5.7 ± 3.9 55.4 ± 28.1
4.7 ± 2.7 43.1 ± 16.1
0.55 0.29
BrS, Brugada syndrome; BrP, Brugada phenocopy; base, base of the triangle. p values of ≤ 0.05 were considered significant.
4
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
Fig. 2. Type 2 Brugada ECG pattern from a patient with true BrS as confirmed by provocative testing (panel A). Type 2 Brugada ECG pattern from a patient with BrP in the context of hyperkalemia (panel B).
Limitations The criteria described by Chevallier et al. and Serra et al. were only validated on type 2 Brugada ECG patterns and therefore our study excluded all type 1 BrP [9,10]. In addition, at the time of the study, there were few confirmed type 2 BrP registered in the international registry [3]. A single operator has performed all measurements. The presence of a second operator would allow for kappa inter-observer analysis. Base of the triangle is reported in mm rather than ms to be consistent with prior literature. This would not allow extrapolation to ECGs recorded at 50 mm/s in which case, transformation to a cut-off of 160 ms is required. Finally, of the applicable publications of BrP, there were a number that had to be excluded because the quality of the ECGs was too low for interpretation. Therefore one limitation of this study is the small number of cases analyzed. Larger studies with increased sample sizes would provide stronger data. As the registry grows, it will allow for larger analytical studies. Conclusions Our study provides evidence that the ECG patterns of BrS and BrP are not reliably distinguishable through employment of these new methodologies when presenting with type-2 ECG Brugada pattern. It provides a second layer of evidence
Table 3 Sensitivity, specificity, predictive values, and likelihood ratios. Measurement
Sensitivity Specificity PPV
NPV
LR + LR −
Base β-angle Base and β-angle Base or β-angle
80% 60% 60% 80%
66.7% 63.6% 63.6% 66.7%
1.4 2.7 2.7 1.4
44% 77.8% 77.8% 44%
61.5% 75% 75% 61.5%
0.5 0.5 0.5 0.5
PPV, positive predictive values; NPV, negative predictive value; LR +, positive likelihood ratio; LR −, negative likelihood ratio; base, base of the triangle.
that the two patterns are identical. In cases where the diagnosis is equivocal, our findings highlight the importance of applying a systematic diagnostic approach to differentiating BrP from BrS. Making a diagnosis on the basis of ECG interpretation alone may negatively impact treatment decisions and potentially have implications on patient morbidity and mortality. Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jelectrocard.2015.12.011.
References [1] Anselm DD, Baranchuk A. Brugada phenocopy: redefinition and updated classification. Am J Cardiol 2013;111:453. [2] Baranchuk A, Nguyen T, Ryu MH, Femenia F, Zareba W, Wilde AA, et al. Brugada phenocopy: new terminology and proposed classification. Ann Noninvasive Electrocardiol 2012;17:299–314. [3] Gottschalk BH, Anselm DD, Baranchuk A. Brugada phenocopy international registry and online educational porta. Available at: www.brugadaphenocopy.com [Accessed September 10, 2015]. [4] Taggart NW, Haglund CM, Tester DJ, Ackerman MJ. Diagnostic miscues in congenital long-QT syndrome. Circulation 2007;115:2613–20. [5] Anselm DD, Gottschalk BH, Baranchuk A. Brugada phenocopies: consideration of morphologic criteria and early findings from an international registry. Can J Cardiol 2014;30:1511–5. [6] Postema PG, Wolpert C, Amin AS, Probst V, Borggrefe M, Roden DM, et al. Drugs and Brugada syndrome patients: review of the literature, recommendations, and an up-to-date website (www.brugadadrugs.org). Heart Rhythm 2009;6:1335–41. [7] Bayés de Luna A, Brugada J, Baranchuk A, Borggrefe M, Breithardt G, Goldwasser D, et al. Current electrocardiographic criteria for diagnosis of Brugada pattern: a consensus report. J Electrocardiol 2012;45:433–42. [8] Gottschalk BG, Anselm DD, Brugada J, Brugada P, Wilde AA, Chiale PA, et al. Expert cardiologists cannot distinguish between Brugada phenocopy and Brugada syndrome ECG patterns. Europace 2015 [in press]. [9] Chevallier S, Forclaz A, Tenkorang J, Ahmad Y, Faouzi M, Graf D, et al. New electrocardiographic criteria for discriminating between Brugada types 2 and 3 patterns and incomplete right bundle branch block. J Am Coll Cardiol 2011;58:2290–8. [10] Serra G, Baranchuk A, Bayés-De-Luna A, Brugada J, Goldwasser D, Capulzini L, et al. New electrocardiographic criteria to differentiate the type-2 Brugada pattern from electrocardiogram of healthy athletes with r′-wave in leads V1/V2. Europace 2014;16:1639–45. [11] Kurisu S, Inoue I, Kawagoe T. Brugada-like electrocardiographic pattern due to hyperkalemia. Clin Cardiol 2009;32:E23.
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx [12] Wolpert C, Echternach C, Veltmann C, Antzelevitch C, Thomas GP, Spehl S, et al. Intravenous drug challenge using flecainide and ajmaline in patients with Brugada syndrome. Heart Rhythm 2005;2:254–60. [13] Wilde AA, Antzelevitch C, Borggrefe M, Brugada J, Brugada R, Brugada P, et al. Proposed diagnostic criteria for the Brugada syndrome: consensus report. Circulation 2002;106:2514–9. [14] Anselm DD, Barbosa-Barros R, de Sousa Belém L, Nogueira de Macedo R, Pérez-Riera AR, Baranchuk A. Brugada phenocopy induced by acute inferior ST-segment elevation myocardial infarction with right ventricular involvement. Inn Card Rhythm Manag 2013;4:1092–4.
5
[15] Gottschalk BH, Anselm DD, Baranchuk A. Brugada phenocopy induced by ischemia or Brugada syndrome unmasked by ischemia? Int J Cardiol 2014;177:619–20. [16] Gottschalk BH, Anselm DD, Baranchuk A. Brugada syndrome unmasked by ischemia needs full risk evaluation. Intern Emerg Med 2015;10:111–2. [17] Recasens L, Merono O, Ribas N. Hyperkalemia mimicking a pattern of Brugada syndrome. Rev Esp Cardiol 2013;66:309. [18] Postema PG, Vlaar AP, DeVries JH, Tan HL. Familial Brugada syndrome uncovered by hyperkalaemic diabetic ketoacidosis. Europace 2011;13:1509–10.
ScienceDirect Journal of Electrocardiology xx (2015) xxx – xxx www.jecgonline.com
New methodologies for measuring Brugada ECG patterns cannot differentiate the ECG pattern of Brugada syndrome from Brugada phenocopy☆,☆☆ Byron H. Gottschalk, BMSc, a Javier Garcia-Niebla, RN, b Daniel D. Anselm, MD, c Atul Jaidka, BMSc, a Antoni Bayés De Luna, MD PhD, d Adrian Baranchuk, MD, FACC, FRCPC a,⁎ a
Department of Cardiology, Kingston General Hospital, Queen's University, Kingston, ON, Canada Servicios Sanitarios del Área de Salud de El Hierro, Valle del Golfo Health Center, Canary Island, Spain Libin Cardiovascular Institute of Alberta, Foothills Medical Centre, University of Calgary, Calgary, AB, Canada d Cardiovascular Research Center, CSIC-ICCC, Hospital de la Santa CreuiSant Pau, Barcelona, Spain b
c
Abstract
Background: Brugada phenocopies (BrP) are clinical entities characterized by ECG patterns that are identical to true Brugada syndrome (BrS), but are elicited by various clinical circumstances. A recent study demonstrated that the patterns of BrP and BrS are indistinguishable under the naked eye, thereby validating the concept that the patterns are identical. Objective: The aim of our study was to determine whether recently developed ECG criteria would allow for discrimination between type-2 BrS ECG pattern and type-2 BrP ECG pattern. Methods: Ten ECGs from confirmed BrS (aborted sudden death, transformation into type 1 upon sodium channel blocking test and/or ventricular arrhythmias, positive genetics) cases and 9 ECGs from confirmed BrP were included in the study. Surface 12-lead ECGs were scanned, saved in JPEG format for blind measurement of two values: (i) β-angle; and (ii) the base of the triangle. Cut-off values of ≥ 58° for the β-angle and ≥ 4 mm for the base of the triangle were used to determine the BrS ECG pattern. Results: Mean values for the β-angle in leads V1 and V2 were 66.7 ± 25.5 and 55.4 ± 28.1 for BrS and 54.1 ± 26.5 and 43.1 ± 16.1 for BrP respectively (p = NS). Mean values for the base of the triangle in V1 and V2 were 7.5 ± 3.9 and 5.7 ± 3.9 for BrS and 5.6 ± 3.2 and 4.7 ± 2.7 for BrP respectively (p = NS). The β-angle had a sensitivity of 60%, specificity of 78% (LR + 2.7, LR − 0.5). The base of the triangle had a sensitivity of 80%, specificity of 40% (LR + 1.4, LR − 0.5). Conclusions: New ECG criteria presented relatively low sensitivity and specificity, positive and negative predictive values to discriminate between BrS and BrP ECG patterns, providing further evidence that the two patterns are identical. © 2015 Elsevier Inc. All rights reserved.
Keywords:
Brugada ECG pattern; Brugada phenocopy; Brugada syndrome; ECG criteria; Beta angle; Base of triangle
Introduction Brugada phenocopies (BrP) are clinical entities characterized by ECG patterns presumed to be identical to those of true Brugada syndrome (BrS) but are elicited by various other underlying conditions [1–3]. Differentiation between the two conditions is of the utmost importance as the ☆
Conflict of Interest Disclosure: None. Funding: None. ⁎ Corresponding author at: Heart Rhythm Service, Cardiac Electrophysiology and Pacing, Kingston General Hospital, Queen's University, K7L 2V7 Kingston, ON, Canada. E-mail address: [email protected] ☆☆
http://dx.doi.org/10.1016/j.jelectrocard.2015.12.011 0022-0736/© 2015 Elsevier Inc. All rights reserved.
treatment recommendations of BrP differ from BrS and misdiagnosis may lead to significant morbidity and mortality [4]. While individuals with BrS are at risk for sudden cardiac death (SCD), and may be candidates for an implantable cardioverter-defibrillator (ICD), the clinical implications of BrP remain unknown. Moreover, patients with true congenital BrS should refrain from using drugs with sodium channel blocking properties, as these have been shown to increase the risk of developing potentially fatal arrhythmias [5–7]. Whether the ECG pattern in BrP serves as a warning of increased arrhythmogenicity has yet to be determined. Therefore, current recommendations for the treatment of BrP focus on resolution of the underlying condition [5].
2
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
Table 1 Brugada phenocopy systematic diagnostic criteria (reproduced with permission [5]). Brugada phenocopy diagnostic criteria (first 5 mandatory) i. ii. iii. iv.
ECG pattern has type 1 or type 2 Brugada morphologic characteristics Patient has underlying condition that is identifiable ECG pattern resolves after resolution of the underlying condition There is low clinical pretest probability of true Brugada syndrome determined by lack of symptoms, medical history, and family history v. Negative results on provocative testing with sodium channel blockers such as ajmaline, flecainide, or procainamide (unless clearly not clinically indicated) vi. Provocative testing not mandatory if surgical right ventricular outflow tract manipulation has occurred within last 96 h vii. Results of genetic testing are negative (desirable but not mandatory because the SCN5A mutation is identified in only 20%–30% of probands affected by true Brugada syndrome) Features that suggest true congenital Brugada syndrome i. ECG pattern shows type 1 or type 2 Brugada morphologic characteristics ii. There is a high clinical pretest probability of true congenital Brugada syndrome determined by presence of symptoms, medical history, and family history iii. Positive results on provocative testing with sodium channel blockers such as ajmaline, flecainide, or procainamide; this indicates sodium channel dysfunction consistent with true Brugada syndrome iv. Genetic testing is positive in about 20%–30% of probands ECG, electrocardiographic; SCN5A, sodium channel voltage-gated type V alpha subunit. Adapted from Anselm et al [5].
Diagnosis of BrP relies on a series of electrocardiographic and clinical characteristics that distinguishes the condition from BrS (Table 1). The first mandatory criterion requires an ECG pattern that is identical to the type 1 or type 2 Brugada ECG pattern. However, until recently, this concept had never been validated. A recent study demonstrated that the patterns of BrP and BrS are indistinguishable under the naked eye [8]. The objective of this study was to determine whether measuring the β-angle [9] and the base of the triangle [10] would allow for a distinction between the type 2 BrP and type 2 BrS pattern.
Methods Ten ECGs from confirmed BrS cases (aborted sudden death, transformation into type 1 upon sodium channel blocking test and/or ventricular arrhythmias, positive genetics) all presenting with type 2 ECG Brugada pattern and 9 ECGs from confirmed BrP (again, all presenting type 2 ECG Brugada pattern) were included in the study. Surface 12-lead ECGs were scanned, saved in JPEG format at 150– 300 DPI, and sent to an evaluator (JGN) in randomized order for evaluation, labeled with only a case number. Images were entered into Autodesk AutoCAD® 2013 software for Windows (San Rafael, CA). Once calibrated, two variables were measured: (i) β-angle, defined as the angle between the upslope of the S-wave and the downslope of the r′-wave (Fig. 1A); and (ii) the base of the triangle, defined as the duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV from the high take-off (Fig. 1B). Three complexes were measured in leads V1 and V2 for each case. These values were averaged to give 1 value for each measured variable in leads V1 and V2. Diagnosis was determined completely by measurement, with no clinical history provided. Cut-off values of ≥ 58° for the β-angle and ≥ 4 mm for the base of the triangle were used to determine the Brugada ECG pattern. A case met criteria for a Brugada ECG pattern if either V1 or V2 (or both) exceeded the cut-off values for the variable being measured. Results are described as mean ± SD for continuous variables and percentage (%) for categorical variables. P values b 0.05 were considered statistically significant. Sensitivity, specificity and likelihood ratios for each variable were calculated to determine diagnostic accuracy. Statistical analysis was performed on IBM SPSS software, version 22 for Windows (Armonk, NY, 2014).
Results The average value for the β-angle was 61.3 ± 25.7° in V1 for all cases and 49.3 ± 23.1° in V2 for all cases. The
Fig. 1. Panel A: The β-angle is calculated by measuring the angle between the upslope of the S-wave and the downslope of the r′-wave. Panel B: The base of the triangle is calculated by measuring duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV from the high take-off.
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
average value for the β-angle in V1 was 66.7 ± 25.5° within the BrS group, and 54.1 ± 26.5° within the BrP group (p = 0.85). The average value for the β-angle in V2 was 55.4 ± 28.1° within the BrS group, and 43.1 ± 16.1° within the BrP group (p = 0.29). The average measurement of the base of the triangle was 6.6 ± 3.6 mm in V1 and 5.3 ± 3.3 mm for all cases. The base of the triangle measured on average 7.5 ± 3.9 mm in lead V1 within the BrS group and 5.6 ± 3.2 mm in V1 within the BrP group (p = 0.85). Similarly, the average measurement of the base of the triangle in lead V2 was 5.7 ± 3.9 mm within the BrS group and 4.7 ± 2.7 mm within the BrP group (p = 0.55). With a cut-off of 58°, use of the β-angle to determine the correct diagnosis had a sensitivity of 60%, specificity of 78% with a corresponding positive likelihood ratio of 2.7 and negative likelihood ratio of 0.5. Using a cut-off value of 4 mm, measuring the base of the triangle to determine the correct diagnosis had a sensitivity of 80% and a specificity of 40% with a positive likelihood ratio of 1.4 and a negative likelihood ratio of 0.5 (Table 2).
3
which had a respective sensitivity and specificity of 95.6% and 85%, for the detection of Brugada ECG pattern. Our data suggest that the new methodologies, previously shown to accurately distinguish between a Brugada ECG pattern and patterns of similar morphology (IRBBB or healthy athletes [9,10]), are unable to distinguish a BrS ECG pattern from that of BrP (when both present with type 2 ECG Brugada pattern) (Fig. 2). The difference in the average values for the β-angle and the base of the triangle were not statistically significant between the two groups. Out of the two variables measured, the base of the triangle provided the greatest sensitivity in identifying a Brugada ECG pattern; however the specificity was poor in excluding the BrP ECG patterns. These data suggest a poor ability of the base of the triangle in distinguishing between the ECG patterns of BrS and BrP. While measurements of the β-angle provided an improved specificity when compared to the base of the triangle, the sensitivity was relatively low. Using the β-angle, 11 of the 19 ECG patterns were diagnosed as a non-Brugada ECG pattern, with 4 of the 11 (36%) cases being true BrS. Interestingly; there were no cases that met the criteria for a diagnosis of a Brugada ECG pattern using the β-angle that did not meet the criteria using the base of the triangle. Therefore, when combining both variables in the analysis, there was no improvement of the sensitivity or specificity (Table 3). The critical findings in this study are that neither the base of the triangle nor the β-angle, or any combination of the two variables were able to reliably and accurately distinguish between the ECG patterns of BrP and BrS (Fig. 2). Additionally, the low sensitivity, specificity and weak likelihood ratios provide further evidence that the diagnostic accuracy of BrP and BrS based only on ECG criteria alone are poor. Our data expand on recent evidence that BrP and BrS ECG patterns are identical and indistinguishable [8]. This information reaffirms the notion that systematic diagnostic criteria are indeed required (Table 1). Perhaps the most important component in the reliable distinction between BrS and BrP is the use of a sodium channel blocking agent in a provocative test. Ajmaline has been shown to be more effective than flecainide in the induction or enhancement of the type 1 Brugada ECG pattern [12]. It is therefore recommended that ajmaline be used in these situations whenever possible. Should ajmaline be unavailable, we suggest the use of flecainide as the provocative drug of choice as it has been shown to be more effective than procainamide [13]. This testing is especially important in situations where a clinical condition has been shown to cause BrP as well as unmask BrS such as myocardial ischemia or hyperkalemia [11,14–18].
Discussion In 2011, Chevallier et al. [9] applied newly developed ECG criteria to successfully distinguish between type-2 Brugada ECG pattern and incomplete right bundle branch block (IRBBB) ECG pattern. They measured values of two ECG parameters: (i) the angle between a vertical line and the downslope of the r′-wave (α-angle); and (ii) the angle between the upslope of the S-wave and the downslope of the r′-wave (β-angle) (Fig. 1A) (see Movie 1 in Supplementary Material) [9]. The authors determined that cut-off values of the α- and β-angles at ≥ 50 and ≥ 58° respectively were optimal for the detection of Brugada ECG pattern. Subsequently, Serra et al. [10] validated the use of the β-angle along with three new criteria in the distinction of the ECG of healthy athletes with r′-waves from type-2 Brugada ECG pattern. The three new criteria included: (i) the duration of the base of the triangle between the upslope and the downslope of the r′-wave at 0.5 mV (5 mm) from the high take-off; (ii) the duration of the base of the triangle at the isoelectric line; and (iii) the ratio of base/height of the triangle formed by the upslope and downslope of the r′-wave at the isoelectric line (Fig. 1B) (see Movie 2 in Supplementary Material) [10]. The easiest parameter to measure from the criteria described by Serra et al. was the duration of the base of the triangle at 5 mm from the high take-off. They determined an optimal cut-off value of ≥ 160 ms (4 mm),
Table 2 Average measurements for BrS compared to BrP. Measurement
BrS V1
BrP V1
P value
BrS V2
BrP V2
P value
Base β-angle
7.5 ± 3.9 66.7 ± 25.5
5.6 ± 3.2 54.1 ± 26.5
0.85 0.85
5.7 ± 3.9 55.4 ± 28.1
4.7 ± 2.7 43.1 ± 16.1
0.55 0.29
BrS, Brugada syndrome; BrP, Brugada phenocopy; base, base of the triangle. p values of ≤ 0.05 were considered significant.
4
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx
Fig. 2. Type 2 Brugada ECG pattern from a patient with true BrS as confirmed by provocative testing (panel A). Type 2 Brugada ECG pattern from a patient with BrP in the context of hyperkalemia (panel B).
Limitations The criteria described by Chevallier et al. and Serra et al. were only validated on type 2 Brugada ECG patterns and therefore our study excluded all type 1 BrP [9,10]. In addition, at the time of the study, there were few confirmed type 2 BrP registered in the international registry [3]. A single operator has performed all measurements. The presence of a second operator would allow for kappa inter-observer analysis. Base of the triangle is reported in mm rather than ms to be consistent with prior literature. This would not allow extrapolation to ECGs recorded at 50 mm/s in which case, transformation to a cut-off of 160 ms is required. Finally, of the applicable publications of BrP, there were a number that had to be excluded because the quality of the ECGs was too low for interpretation. Therefore one limitation of this study is the small number of cases analyzed. Larger studies with increased sample sizes would provide stronger data. As the registry grows, it will allow for larger analytical studies. Conclusions Our study provides evidence that the ECG patterns of BrS and BrP are not reliably distinguishable through employment of these new methodologies when presenting with type-2 ECG Brugada pattern. It provides a second layer of evidence
Table 3 Sensitivity, specificity, predictive values, and likelihood ratios. Measurement
Sensitivity Specificity PPV
NPV
LR + LR −
Base β-angle Base and β-angle Base or β-angle
80% 60% 60% 80%
66.7% 63.6% 63.6% 66.7%
1.4 2.7 2.7 1.4
44% 77.8% 77.8% 44%
61.5% 75% 75% 61.5%
0.5 0.5 0.5 0.5
PPV, positive predictive values; NPV, negative predictive value; LR +, positive likelihood ratio; LR −, negative likelihood ratio; base, base of the triangle.
that the two patterns are identical. In cases where the diagnosis is equivocal, our findings highlight the importance of applying a systematic diagnostic approach to differentiating BrP from BrS. Making a diagnosis on the basis of ECG interpretation alone may negatively impact treatment decisions and potentially have implications on patient morbidity and mortality. Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jelectrocard.2015.12.011.
References [1] Anselm DD, Baranchuk A. Brugada phenocopy: redefinition and updated classification. Am J Cardiol 2013;111:453. [2] Baranchuk A, Nguyen T, Ryu MH, Femenia F, Zareba W, Wilde AA, et al. Brugada phenocopy: new terminology and proposed classification. Ann Noninvasive Electrocardiol 2012;17:299–314. [3] Gottschalk BH, Anselm DD, Baranchuk A. Brugada phenocopy international registry and online educational porta. Available at: www.brugadaphenocopy.com [Accessed September 10, 2015]. [4] Taggart NW, Haglund CM, Tester DJ, Ackerman MJ. Diagnostic miscues in congenital long-QT syndrome. Circulation 2007;115:2613–20. [5] Anselm DD, Gottschalk BH, Baranchuk A. Brugada phenocopies: consideration of morphologic criteria and early findings from an international registry. Can J Cardiol 2014;30:1511–5. [6] Postema PG, Wolpert C, Amin AS, Probst V, Borggrefe M, Roden DM, et al. Drugs and Brugada syndrome patients: review of the literature, recommendations, and an up-to-date website (www.brugadadrugs.org). Heart Rhythm 2009;6:1335–41. [7] Bayés de Luna A, Brugada J, Baranchuk A, Borggrefe M, Breithardt G, Goldwasser D, et al. Current electrocardiographic criteria for diagnosis of Brugada pattern: a consensus report. J Electrocardiol 2012;45:433–42. [8] Gottschalk BG, Anselm DD, Brugada J, Brugada P, Wilde AA, Chiale PA, et al. Expert cardiologists cannot distinguish between Brugada phenocopy and Brugada syndrome ECG patterns. Europace 2015 [in press]. [9] Chevallier S, Forclaz A, Tenkorang J, Ahmad Y, Faouzi M, Graf D, et al. New electrocardiographic criteria for discriminating between Brugada types 2 and 3 patterns and incomplete right bundle branch block. J Am Coll Cardiol 2011;58:2290–8. [10] Serra G, Baranchuk A, Bayés-De-Luna A, Brugada J, Goldwasser D, Capulzini L, et al. New electrocardiographic criteria to differentiate the type-2 Brugada pattern from electrocardiogram of healthy athletes with r′-wave in leads V1/V2. Europace 2014;16:1639–45. [11] Kurisu S, Inoue I, Kawagoe T. Brugada-like electrocardiographic pattern due to hyperkalemia. Clin Cardiol 2009;32:E23.
B.H. Gottschalk et al. / Journal of Electrocardiology xx (2015) xxx–xxx [12] Wolpert C, Echternach C, Veltmann C, Antzelevitch C, Thomas GP, Spehl S, et al. Intravenous drug challenge using flecainide and ajmaline in patients with Brugada syndrome. Heart Rhythm 2005;2:254–60. [13] Wilde AA, Antzelevitch C, Borggrefe M, Brugada J, Brugada R, Brugada P, et al. Proposed diagnostic criteria for the Brugada syndrome: consensus report. Circulation 2002;106:2514–9. [14] Anselm DD, Barbosa-Barros R, de Sousa Belém L, Nogueira de Macedo R, Pérez-Riera AR, Baranchuk A. Brugada phenocopy induced by acute inferior ST-segment elevation myocardial infarction with right ventricular involvement. Inn Card Rhythm Manag 2013;4:1092–4.
5
[15] Gottschalk BH, Anselm DD, Baranchuk A. Brugada phenocopy induced by ischemia or Brugada syndrome unmasked by ischemia? Int J Cardiol 2014;177:619–20. [16] Gottschalk BH, Anselm DD, Baranchuk A. Brugada syndrome unmasked by ischemia needs full risk evaluation. Intern Emerg Med 2015;10:111–2. [17] Recasens L, Merono O, Ribas N. Hyperkalemia mimicking a pattern of Brugada syndrome. Rev Esp Cardiol 2013;66:309. [18] Postema PG, Vlaar AP, DeVries JH, Tan HL. Familial Brugada syndrome uncovered by hyperkalaemic diabetic ketoacidosis. Europace 2011;13:1509–10.