Available online at www.sciencedirect.com

Journal of Electrocardiology 45 (2012) 433–442www.jecgonline.com

Current electrocardiographic criteria for diagnosis of Brugada pattern: aconsensus report☆

Antonio Bayés de Luna, MD, PhD, a,⁎ Josep Brugada, MD, PhD, b Adrian Baranchuk, MD, c

Martin Borggrefe, MD, d Guenter Breithardt, MD, e Diego Goldwasser, MD, a

Pier Lambiase, MD, f Andrés Pérez Riera, MD, PhD, g Javier Garcia-Niebla, RN, h

Carlos Pastore, MD, PhD, i Giuseppe Oreto, MD, j William McKenna, MD, f

Wojciech Zareba, MD, PhD, k Ramon Brugada, MD, PhD, l Pedro Brugada, MD, PhDm

aInstitut Català Ciències Cardiovasculars–Hospital Sant Pau, Barcelona, SpainbHospital Clínico, Barcelona, Spain

cKingston General Hospital, Kingston, CanadadUniversitätsmedizin Mannheim-I. Medizinische Klinik, Mannheim, Germany

eMed.Klinik & Poliklinik, Universität Münster, Münster, GermanyfHeart Hospital, UCL, London, UK

gFacultad de Medicina del ABC-Santo André, Sao Paulo, BrasilhCannary Islands, Spain

iInstituto do Coraçao, Sao Paulo, BrasiljUniversita de Messina, Messina, Italy

kUniversity of Rochester Medical Center, Rochester, NY, USAlCardiovascular Genetic Center UdG-IDIBGI, Girona, Spain

mFree University of Brussels (UZ Brussel) VUB, Brussels, Belgium

Received 20 March 2012

Abstract Brugada syndrome is an inherited heart disease without structural abnormalities that is thought to

☆ Sponsored by tElectrocardiology (ISH

⁎ CorrespondingCatalà Ciències CardBarcelona, Spain.

E-mail addresses:

0022-0736/$ – see frodoi:10.1016/j.jelectroc

arise as a result of accelerated inactivation of Na channels and predominance of transient outwardK current (Ito) to generate a voltage gradient in the right ventricular layers. This gradient triggersventricular tachycardia/ventricular fibrillation possibly through a phase 2 reentrant mechanism.The Brugada electrocardiographic (ECG) pattern, which can be dynamic and is sometimesconcealed, being only recorded in upper precordial leads, is the hallmark of Brugada syndrome.Because of limitations of previous consensus documents describing the Brugada ECG pattern,especially in relation to the differences between types 2 and 3, a new consensus report to establisha set of new ECG criteria with higher accuracy has been considered necessary. In the new ECGcriteria, only 2 ECG patterns are considered: pattern 1 identical to classic type 1 of otherconsensus (coved pattern) and pattern 2 that joins patterns 2 and 3 of previous consensus (saddle-back pattern). This consensus document describes the most important characteristics of 2 patternsand also the key points of differential diagnosis with different conditions that lead to Brugada-likepattern in the right precordial leads, especially right bundle-branch block, athletes, pectusexcavatum, and arrhythmogenic right ventricular dysplasia/cardiomyopathy.Also discussed is the concept of Brugada phenocopies that are ECG patterns characteristic ofBrugada pattern that may appear and disappear in relation with multiple causes but are not relatedwith Brugada syndrome.© 2012 Elsevier Inc. All rights reserved.

Keywords: r' in V1; ST elevation; Brugada syndrome

he International Society for Holter and NoninvasiveNE).

author. Universidad Autónoma de Brcelona, Institutiovasculars, S.Antoni M. Claret 167, ES 08025

abayes@csic-iccc.org, msauri@csic-iccc.org

nt matter © 2012 Elsevier Inc. All rights reserved.ard.2012.06.004

Concept

Brugada syndrome (BrS)1 is a familial, geneticallydetermined syndrome characterized by autosomal dominantinheritance in about 50% of cases with variable penetrance.

Fig. 1. In BrS, the heterogeneous dispersion of repolarization takes placesbetween the endocardium and the epicardium of the RV at the beginning ofphase 2 (voltage gradient [VG]) because of the transient predominance ofoutward Ito current. Epi indicates epicardium; Endo, endocardium.

434 A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

More than 70 mutations, most commonly in the cardiac Na+

channel, have been described. In around 20% of cases, thereare SCN5A mutations that explain the accelerated inactiva-tion of Na channels.2

The diagnosis of BrS is suggested by the clinical historyin a patient with specific electrocardiographic (ECG) pattern(Brugada pattern [BrP]) (Table 2). Sometimes, it is necessaryto use other electrocardiological findings and methods(Table 1B and C).

The number of idiopathic ventricular fibrillation (VF)cases diagnosed as having BrS depends on the ECGdiagnostic criteria used. In fact, the ECG manifestation(BrP), although crucial, is only a part of the BrS diagnosticcriteria. The ECG criteria are a key point not only fordiagnosis but also for prognosis and risk stratification.However, we do not want to discuss all these aspects in thisarticle as we have stated. We plan only to comment on the12-lead ECG clues for the diagnosis of BrS and how toperform the differential diagnosis with other ECG patternsthat present with ST elevation in V1-V2 and/or r' in these

Table 1Electrocardiogram alterations in Brugada syndrome

A. ECG diagnostic criteriaChanges in precordial leads.1. Morphology of QRS-T in V1-V3. ST elevation (sometimes only in V1

and exceptionally also in V3) (BrP)Type 1. Coved pattern: initial ST elevation ≥2 mm, slowly descending and

concave or rectilinear with respect to the isoelectric baseline, withnegative symmetric T wave (see other characteristics in Table 2 and text).

Type 2. Saddle back pattern: The high take-off (r`) is≥2 mm with respect tothe isoelectric line and is followed by ST elevation; convex with respect tothe isoelectric baseline with elevation ≥0.05 mV with positive/flat Twave in V2 and T wave variable in V1. If there is some doubt (ie, r' b2mm), it is necessary to record the ECG in 2nd, 3rd ICS. Othercharacteristics may be seen in Table 2 and text.

2. New ECG criteria:a. Corrado index (2010): Ratio high take-off of QRS-ST/height of ST at 80

ms later is in V1-V2 N1 because the ST is downsloping. In athletes, the STespecially in V2 is upsloping; and the index is b1. The end of QRS (Jpoint) often does not coincide with the high take-off of QRS-ST as wassuggested by Corrado.36 However, using the high take-off of QRS-ST forthe application of the Corrado index is valid for discriminating BrP andother conditions mimicking BrP.

b. The β angle formed by ascending S and descending r' is N58º in type 2BrP (in athletes, it is much lower) (Chevalier et al. 2011) (SE 79%SP:84%).

c. Duration of the base triangle of r' at 5 mm from the high take-off is morethan 3.5 mm in BrP 2 (SE, 81%; SP, 82%).31

B. Other ECG findings1. QT generally is normal. May be prolonged in right precordial leads.43

2. Conduction disorders: Sometimes, prolonged PR interval (longHV interval).The conduction delay located in RV explains the r' and longer QRS durationin right precordial leads compared with mid/left precordial leads.

3. Supraventricular arrhythmias. Mostly atrial fibrillation.4. Some other ECG findings may be seen: the presence of r' wave in aVR N3

mm,44 early repolarization pattern in inferior leads,45 fractioned QRS,46

alternans of T wave after ajmaline injection,47 etc.

C. Other electrocardiological techniques: In some occasions, it is necessaryor convenient to find some new diagnostic clues with exercise testing,48

late potentials,49 and impaired QT dynamics studied by Holter.Electrophysiological studies remains controversial for diagnosis and forrisk stratification50–52

leads. In the presence of the ECG criteria of BrP, BrS isdiagnosed if one/more of the following clinical factors arepresent: (a) survivors of cardiac arrest, (b) presence ofpolymorphic ventricular tachycardia (VT), (c) history ofnonvagal syncope, (d) familial antecedents of sudden deathin patients younger than 45 years without acute coronarysyndrome, and (e) or type 1 ST pattern in relatives.

The prevalence of the symptoms is estimated to be 5 per10000 inhabitants in Southeast Asia; and apart fromaccidents, it is the leading cause of death in men youngerthan 40 years in this part of the world.3

The incidence is higher in young men without apparentstructural heart disease, but this concept of the structurallynormal heart in BrS has been challenged.4 Syncope orsudden death usually occurs during rest or sleep, sometimeswithout any isolated, preceding premature ventricularcontractions. Most often, polymorphic ventricular tachycar-dia triggers ventricular fibrillation possibly because of aphase 2 reentrant mechanism due to a voltage gradientbetween the different action potential duration of layers ofthe right ventricle (Fig. 1).

There are currently 2 pathophysiologic hypotheses usedto explain the ECG changes in BrS5: (a) the repolarizationhypothesis,6 involving the presence of a voltage gradient dueto transmural or transregional dispersion of the actionpotential of different layers of the right ventricle at thebeginning of repolarization, as a consequence of a loss of Nacurrent combined with a dominant transient outward Kcurrent (Ito) (Fig. 1) and (ii) the depolarization hypothesis,involving right ventricular conduction delay at the end ofdepolarization7 combined probably with subtle structuralright ventricular anomalies.8 This latter mechanism isindirectly supported by the presence of late potentials,9

fragmented potentials on the anterior aspect of rightventricular outflow tract (RVOT),10 and delayed rightventricle free wall contraction after ajmaline infusion.11

Borggrefe and Schimpf12 commented that both mecha-nisms may play a role in the genesis of the ECG pattern andthat both may coexist. Elizari et al 13 published thespeculative but fascinating theory encompassing the 2

435A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

previous ones proposing that abnormal neural crest cells inthe RVOT cause conduction delay, resulting in the typicalST changes of ECG BrP. The repolarization gradients mayoccur not only between epicardium and endocardium butalso between RVOT and normal surrounding myocardium.In fact, the evidence of right ventricular conduction delayand abnormal repolarization, which are already seen in theECG (at least part of the QRS end and symmetric T wave),are also demonstrated by VCG (delay of final part of QRSloop and rounded T loop).14 Also Hoogendijk et al,15 basedon the theory that explains ST-segment elevation due toexcitation failure by current-to-load mismatch, proposed aunifying mechanism for the Brugada pattern. Although boththese unifying theories are scientifically attractive, we needfurther studies to definitively establish the underlyingmechanism of BrS.16 It is likely that both mechanisms, theRVOT conduction delay and the voltage gradient duringphase 2 of repolarization, play a role in the explanation ofECG pattern and in the triggering of VT/VF.

ECG abnormalities of the BrP

Firstly, we have to say that the abnormal ECG constitutesthe hallmark of BrS; but it is necessary to emphasize that theECG changes can be dynamic and sometimes are concealed.Indeed, the ECG BrP is sometimes intermittent; and it may beobserved only in certain situations, such as fever, intoxication,vagal stimulation, and electrolyte imbalance (Table 1).17,18

Furthermore, there are some drugs (sodium channels blockers)that may unmask a BrP.19 The ECG pattern characteristic ofBrS that may arise from multiple causes (metabolic disorders,electrocution, ischemia, etc) and disappear upon resolution ofthe injury has been named Brugada phenocopy.20 Although

Fig. 2. See how the devices with non-linear-phase filters may produce, when usingventricular hypertrophy, which mimics BrP. Color illustration online.

this concept has to be confirmed, it is considered that theBrugada phenocopies have a negative pharmacologicalchallenge test result (to Na channel blockade) and a negativegenetic test result for recognized BrS mutations.20 Theprognosis in these cases is unknown. For more details, consultwww.brugadadrugs.org.21

It is also important to be sure that the ECG is recordedcorrectly and that there are no artifacts due to the use ofinappropriate filters22 (Fig. 2). However, currently, mostECG devices use linear-phase digital filters that allow cutoffpoints until 0.67 Hz without inducing ST alterations.22,23

Currently, the most commonly used ECG criteria are theones proposed by the 2 consensus documents that werepublished in 2002 and 2005 under the auspices of theEuropean Society of Cardiology.24,25 However, the limita-tions of previous reports, especially in relation to thedifferences between types 2 and 3 ECG patterns, and theevidence that it is very important to define ECG criteriawith a high degree of precision because the decision toimplant an ICD depends on it mean that it is necessary toreview the current criteria. More recently, the criteriaproposed by previous consensus documents have also beenused in other articles related to this topic,26,27 although, ina Japanese study, it was suggested to combine types 2 and3 into only 1 type.28

We will now describe the most useful current ECGcriteria to make this diagnosis and also some other ECGabnormalities that may be found.

New ECG criteria

Morphology in V1-V2Only 2ECGpatterns have to be considered (Tables 1 and 2).

Type 1 is identical to the classic type 1 described in the

high pass filter of 0.5 Hz, changes in the ECG pattern of V2 in case of lef

t

Table 2ECG patterns of BrS in V1-V2

Type 1: coved pattern Type 2: saddle-back pattern

This typical coved pattern present in V1-V2 the following: This typical saddle-back pattern present in V1-V2 the following:a. At the end of QRS, an ascending and quick slope with a high

take-off ≥2mm followed by concave or rectilinear downsloping ST.There are few cases of coved pattern with a hightake-off between 1 and 2 mm.

a. High take-off of r' (that often does not coincide with J point) ≥2 mm.

b. There is no clear r' wave.

b. Descending arm of r' coincides with beginning of ST (often is notwell seen).

c. The high take-off often does not correspond with the J point (Fig. 4 B).

c. Minimum ST ascent ≥0.5 mm

d. At 40 ms of high take-off, the decrease in amplitude of ST is ≤4mm.28

In RBBB and athletes, it is much higher.

d. ST is followed by positiveT wave in V2 (T peak N ST minimum N 0) and of variable morphologyin V1.

e. ST at high take-off N ST at 40 ms N ST at 80 ms.e. The characteristics of triangle formed by r' allow to define differentcriteria useful for diagnosis (see above and text).

f. ST is followed by negative and symmetric T wave • β angle.26

g. The duration of QRS is longer than in RBBB, and there is a mismatchbetween V1 and V6 (see text).

• Duration of the base of the triangle of r' at 5 mm from the high take-offgreater than 3.5mm31

f. The duration of QRS is longer in BrP type 2 than in other cases with r'in V1, and there is a mismatch between V1 and V6 (see text).

436 A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

previous consensus, although we will describe some newparameters to identify it. The new type 2 pattern combinespatterns 2 and 3 of previous consensus. These 2 patterns maybe unified in only one that encompasses both patterns. In fact,

Fig. 3. A, Typical example of BrP type 1. Note the morphology in V1-V2, with awith no clear evidence of R′. B, Typical example of type 2 BrP (saddle-back pat(Table 2).

there are only small morphological differences between the 2of them that do not impact on prognosis and risk stratification.Even with drug challenge,24 conversion of type 3 to type 2 isconsidered inconclusive.

concave ST down-sloping elevation with respect to the isoelectric baseline,tern). Note the morphology in V1-V2 with final r' of special characteristics

437A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

Type 1 (Fig. 3A): This pattern, known as a coved pattern,presents as ST-segment elevation followed by a symmetricnegative T wave in the right precordial leads.1,24,25,28

Usually, the pattern is seen only in V1-V2. But in somecases, it is recorded only in V1 or V2; and, in others, it isobserved from V1 to V3. Usually, a clear r' is not seen. Thistypical electrocardiographic pattern that may be considereddiagnostic of BrP presents the following characteristics(Table 2):

1) The high take-off of QRS-ST that is at least 2 mm in V1is followed by downsloping ST-segment elevationconcave or rectilinear with respect to the isoelectricline. There are a few cases of coved pattern with hightake-off less than 2 mm and at least 1 mm.28 This hightake-off level is higher than ST level after 40 millisec-onds, and this is higher than ST after 80 milliseconds(high take-off N ST after 40 milliseconds N ST after 80milliseconds) (Table 2).

2) At 40 milliseconds from the high take-off of QRS-ST, thedecrease in amplitude is less than 0.4 mV. This is muchless than the decrease observed in right bundle-branchblocks (RBBBs) because the down-slope is slower.28

3) The index QRS-ST elevation at high take-off/height of STat 80 milliseconds later is greater than 1 in BrP and lessthan 1 in athletes (Fig. 6). According to Corrado(Fig. 4A), the high take-off QRS-ST coincides with Jpoint (end of depolarization). However, although thisphenomenon is not present in all cases7,29 (Fig. 4B), thisindex is still valid (Figs. 4 and 6).

4) The QRS duration is longer in V1-V2 than in mid/leftprecordial leads because of evidence of right ventricularconduction delay. However, in many cases, it isimpossible to measure it in the ECG because we cannotidentify exactly the duration of the terminal QRS.Although the VCG may measure with more accuracy

Fig. 4. A, According to Corrado et al,36 the J point coincides with high take-off (seepatient with BrS. See how the J point at the end of QRS in lead II after ajmaline

the milliseconds of the QRS loop, it is also difficult toensure the exact duration of the QRS loop because in thefirst and late part, around 30 to 40 milliseconds, it isslowly inscribed and difficult to measure exactly.

5) The ST segment is followed by an asymmetric T wave.

The ECG BrP type 1 with the above criteria is easilyrecognized; and when found in a patient without apparentstructural heart disease, it strongly supports the diagnosis ofBrS (Tables 1 and 2 and later differential diagnosis). Thecases with coved-type pattern that present a high take-off ofQRS-ST between 0.1 and 0.2 mV but with negative T wavein V1-V2 have been considered suggestive of type 1.28

These cases are very infrequent; and to confirm thediagnosis, some of the other ECG findings listed inTable 1 should be present.

Type 2 (Fig. 3B): It is characterized in V1 and V2 with thepresence of terminal positive wave called r', although in fact,it is a mixture of final QRS and beginning of repolarization.This r' presents a high take-off of at least 0.2 mV ofamplitude, followed by elevated ST segment (≥0.5 mm)convex with respect to the isoelectric line (saddle-backpattern) and by a T wave that is positive in V2 (amplitude ofthe peak of the T N minimum amplitude of ST) and ofvariable morphology in V1: mildly positive, flat, or mildlynegative (Fig. 3B and Table 2). It has been considered thatthe descending limb of the r' terminates when a suddenchange in the slope occurs. However, sometimes, especiallyin V1, there is no clear change in the slope direction so thatthe T wave onset cannot be determined.

In this scenario, it is important to verify whetherintravenous administration of a drug that blocks the sodiumchannels (ajmaline, flecainide, procainamide, pilsicainide)turns the type 2 pattern into type 125,30 (Fig. 5).

The characteristics of the “r'” in BrP are different than inRBBB and other conditions with r' in V1-V2. In a BrP ECG,

text). B, Simultaneous lead ECG recording at baseline and after ajmaline in aoccurs later than the high take-off of QRS-ST junction.7

Fig. 5. A, Electrocardiogram of a young man with nonvasovagal syncope. B, The ECG in fourth intercostal space (ICS) is suspicious (V2). A, With theelectrodes located in second ICS, the ECG was BrP type 2. The diagnosis was confirmed by the ajmaline test, where the typical ECG pattern type 1 wasobserved. The patient experienced ventricular fibrillation with just one extrastimulus in the electrophysiologic study, and an implantable cardioverterdefibrillator was implanted.

438 A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

the high take-off of r' is not peaked; and the descending armof r', considering the “r'” as a triangle, has a gradual slope.The r' of BrP type 2 compared with other types of r' has thefollowing features:

1. The angles between both arms of r´ are wider (α and βangle).26 The sensitivity (SE) and specificity (SP) arehigher in β angle (79% and 84%, respectively) with athreshold of 58º (Table 2).

2. The measurement of the duration of the base of triangle ofr' at 5 mm from the high take-off is more than 3.5 mm inBrP type 2, with an SE of 81% and an SP of 82%.31 Thiscriterion is easier to measure than the α and β angles(Table 2).

Accuracy of these ECG criteria

The use of these ECG criteria is very helpful for thediagnosis of BrP that may be undertaken even bynonexperienced physicians, especially type 1 ECG patternthat has a high specificity.

Recently, in a Japanese population, a computerizeddiagnostic criteria for BrS that demonstrated a high levelof accuracy have been published (type 1 N90%, type 2N85%).28 This study was based on the comparison withRBBB in apparently healthy people but not with athletes andpeople with pectus excavatum.

Other ECG findings or ECG techniques may be used forcorrect diagnosis of BrS

Table 1B and C shows the most important other ECGfindings that may be used for correct diagnosis of BrS.

Changes of morphology according the location ofthe electrodes

In some cases, the ECG pattern is only evident or is moreevident in the upper precordial V1-V2 leads recorded at thethird or second intercostal space. This occurs because theabnormal electrical activity leading to BrP arises from alimited zone located in the RVOT. Accordingly, only anelectrode located exactly over the affected zone is able torecord the BrP. Recently, it has been published32 that, in BrPtype 1, lead positioning according to RVOT location byusing CV magnetic resonance imaging correlations mayimprove the diagnosis of BrP. The ECG pseudonormaliza-tion observed when the electrodes are located in the correctposition does not constitute any guarantee that the diagnosisis incorrect, especially if some ST elevation persists, albeitless markedly (Fig. 5). Therefore, it is very important toposition the electrodes in the same location when serialrecordings are performed to allow comparative assessments,and to perform ECG recordings in upper precordial leads ifsome suggestion of BrP exists in V1-2 in the ECG recordedat the fourth intercostal space, such as minimal ST elevationand/or positive terminal forces of the QRS.

Variability of BrP over time

In most patients, BrP is not evident at any time, changingfrom the most obvious type 1 to a barely recognizable type 2.In a follow-up study based on 90 ICD patients with BrS, onethird of analyzed tracings showed type 1 BrP and one thirdshowed perfectly normal ECG tracings.33 This is notsurprising because the sodium channel dysfunction under-lying BrP is variable, being influenced by different factors(drugs, fever, autonomic imbalance, etc).

439A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

3. Differential diagnosis

We have to distinguish (a) the cases of spontaneoustypical BrP type 1 in case of BrS; (b) the typical BrP type 1induced by some drugs (sodium channel blockers) or othercircumstances (fever, etc) that unmask a BrS; (c) the casesof ECG BrP especially type 1 induced by manycircumstances that disappear upon resolution of the injuryand that do not present BrS (named phenocopies) (Theseinclude acute ischemia, pericarditis, myocarditis, pulmonaryembolism, metabolic disorders, ionic disorders, administra-tion of some drugs, electrocution, and a miscellaneousgroup [consult www.brugada.org and Baranchuk et al20].);and (d) the cases of similar BrP (Brugada-like) that arepermanent patterns and may be confused with a BrP type 1or 2. These include RBBB, septal hypertrophy, arrhythmo-genic right ventricular dysplasia/cardiomiopathy, athletes,pectus excavatum, and a long number of processes (consultwww.brugada.org) (Fig. 6).

Sometimes, the differential diagnosis of BrP from otherconditions associated with ST elevation by ECG alone may bequite difficult, even for the more experienced cardiologist.34

The clinical setting is very important, bearing in mind that, inmany cases, the patient with this problematic pattern presentswith acute symptoms. Patients with acute coronary syndromeswith ST elevation may present in some cases a similar patternbut obviously in a very different clinical setting. In somedoubtful cases of type 2 pattern, to determine the correctdiagnosis, we may check if there are changes after theintravenous administration of Na channel blockers (ajmaline)(already discussed). There are however frequent cases ofBrugada-like pattern, especially type 2, in which the ECG isusually seen in asymptomatic patients that require a detailedECG analysis to obtain the correct interpretation. We willcomment on the most challenging cases (Fig. 6).

Isolated RBBBs

Type 1 BrP and advanced RBBB are both characterizedby a terminal positive wave and a negative T wave in V1-V2.

Fig. 6. Comparison between type 1 and 2 BrP, arrhythmogenic right ventricular dysmismatch in the QRS duration in V1-V2 and mid/left precordial leads. There is clesecond vertical line is located at 80 milliseconds of the J point (first line) to check

The distinction between the 2 conditions is straightforwardbecause, usually, in advanced RBBB, the ST segment isnot elevated in the right precordial leads, the terminalwave (r' or R') is synchronous with the broad S waveobserved in leads I and V6, and the QRS is wider (≥120milliseconds). In type 1 BrP, however, usually no wide Swave is present in the left leads because the terminalforces of the ventricular complex in V1-V2 can berecorded only by electrodes placed in proximity of the sitewhere the abnormal electrical activity occurs (the outflowtract of the right ventricle) and not from further awayleads. Therefore the QRS in left leads is usually is lessthan 120 milliseconds.

Type 2 BrP and incomplete RBBB raise more problems:whenever a small positive terminal QRS deflection (r') ispresent in V1-V2, it is possible to distinguish the r' wave ofincomplete RBBB from the terminal positive r' wave of type2 BrP on the basis of the following: (1) the positive terminalr' wave is peaked in RBBB, whereas, in BrP, type 2 isrounded, wide, and usually of relatively low voltage; (2) inincomplete RBBB, the QRS complex duration in leads V1-V2 is identical to that observed in lead V6. In type 2 BrP, incontrast, the QRS duration is longer in the right precordialleads than in V6 because the terminal deflection observed inV1 (the r' wave) cannot be recorded by the V6 electrode. Inother words, in BrP, the QRS complex end is earlier in V6than in V1-V2.35

Athletes

Type 1 BrP compared with athletes (Fig. 6) depicts an STsegment clearly elevated and down-sloping, usually with novisible r'. This pattern is difficult to be mistaken with theECG pattern seen in V1-V2 in athletes.

In type 2 BrP, the high take-off is an r' usually rounded(see characteristics of r' before), which is followed by adown-sloping ST elevation with a final T wave of variablemorphologies in V1 and positive in V2 (Table 2).

plasia, pectus excavatum, athletes, and partial RBBB to view the presence oar mismatch in type 1 and 2 of BrP, and there is none in the other cases. Thethe Corrado index. Color illustration online.

f

440 A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

Compared with this pattern, the ECG of athletes presentsthe following:

1. The end of the QRS in V1 coincides with the end of theQRS in V5-6, whereas in type 2 BrP, this is not usuallyclearly seen because the QRS complex ends later in V1-V2 than in V6 (see above).

2. In athletes, lead V1 shows an r' usually peaked and sharpwith no ST elevation or only mildly elevated (b1 mm);and the ST segment starts after the clear end of QRS andis often followed by a negative and sometimes deep Twave in V1. Occasionally, the T wave is positive,especially in V2; but in these cases, the ST segment isgradually ascending (Fig. 6). Therefore, the index ratioST-segment elevation at high take-off of QRS-ST/STelevation after 80 milliseconds in V1-V2 distinguishesBrP from athletes: in the latter group, the ratio is less than1, whereas in BrP, it is greater than 1.36

Pectus excavatum

The ECG diagnosis of this normal variant of the ECGseen in pectus excavatum37 is supported by the following:

− A negative P wave in V1 taken in the correct location(fourth ICS)

− A peaked r' in V1 very well defined, which may befollowed by an elevated ST usually mild. The T waveis usually negative or plus/minus in V1 and positive inV2 (Fig. 6).

Arrhythmogenic right ventricular dysplasia/cardiomyopathy

From an electrocardiographic point of view, it is usuallycharacterized by an ECG pattern in V1-V238 different fromBrP by the following:

- Atypical RBBB morphologies have been shown, witha “plateau” in the R wave in lead V1. The ST segmentis at times somewhat elevated but does not usuallymimic type 2 BrP, although, in some cases, it may besimilar because the ε wave may be confused with ther' wave of BrP.

- In arrhythmogenic right ventricular dysplasia/cardio-myopathy, moreover, T waves are more frequentlynegative in many precordial leads (V1 to V3-V5).

The ECG pattern is always fixed.The signal averaging technique may help to differentiate

both processes because although in both of them abnormalhigh-frequency components are present (positive latepotentials), wavelet analysis demonstrates that the frequen-cy level is higher in arrhythmogenic right ventriculardysplasia/cardiomyopathy.39

Early repolarization pattern

The early repolarization variant is a well-recognizedcommon enigmatic idiopathic electrocardiographic phenom-enon, considered to be present when at least 2 adjacentprecordial leads show ST-segment elevation with values of

at least 1 mm. This pattern, which has recently beenassociated to ventricular fibrillation (sudden death),40 maybe defined as a positive, sharp, and well-defined “hump-like”deflection immediately following positive QRS complexesat the onset of the ST segment, recorded especially in themidleft precordial (not right precordial) and sometimesinferior leads. The transition from QRS to ST segment isconsidered “slurred” and not a clear J point (wave) when theR wave gradually becomes the ST segment with uprightconcavity.41 The characteristics of this ECG pattern are verydifferent from BrP, but sometimes both patterns maycoexist.42

Conclusions

We have exposed the current ECG criteria of BrP,including the ones described after the last consensusstatement,24,25 which have very high accuracy and couldused by nonexperienced physicians for a reliable diagnosisof both types of BrP, especially type 1. The most importantpoints are the following:

1. Only 2 ECG patterns should be considered: type 1(coved type) and type 2 (saddle-back type) thatencompass the patterns 2 and 3 of previous consensus(Table 2).

2. In both types 1 and 2, the end of the QRS in themid/left precordial leads (J point) occurs before theend of the QRS in V1 due to localized RVOTconduction delay. However, it is impossible toconfirm by surface ECG the exact difference becauseof the impossibility of determining exactly the end ofQRS in V1. Furthermore, often, the J point in mid/leftprecordial leads does not coincide with the high take-off of the QRS-ST in V1,29 as was considered byCorrado et al.36

3. Type 1 pattern is very specific; and when it appears in apatient without apparent heart disease, it strongly suggeststhe diagnosis. The small number of cases with no typicalpattern (ST elevation b2 mm) has to be evaluated withcomplementary ECG criteria test to reach a final diagnosis(Table 1B and C).

4. Electrocardiographic patterns identical to BrP can be seenin the absence of true BrS (phenocopies).

5. In BrP, the slope of ST in V1-V2 is descending; and inathletes and other cases with r' in V1, it is usuallyascending at least in V2 (Fig. 6).

6. Type 2 pattern may be confused with several othersituations with r' in V1-V2. The characteristics of r' inincomplete RBBB, athletes, and pectus excavatumhowever are especially different from those observed inBrS. These differences include (Table 2) (a) the angle ofascending and descending arms of r'; (b) the duration ofthe base triangle of r' at 5 mm from the high take-off; and(c) the configuration of r' is peaked in RBBB, pectusexcavatum, and athletes and rounded in BrP. The Corradoindex can also be helpful in distinguishing BrP fromathlete's ECG (Table 2 and Fig. 6).

441A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

References

1. Brugada P, Brugada J. Right bundle branch block, persistent STsegment elevation and sudden cardiac death: a distinct clinical andelectrocardiographic syndrome. A multicenter report. J Am Coll Cardiol1992;20:1391.

2. Probst V, Wilde AA, Barc J, et al. SCN5A mutations and the role ofgenetic background in the pathophysiology of Brugada syndrome. CircCardiovasc Genet 2009;2:552.

3. Nademanee K. Sudden unexplained death syndrome in Southeast Asia.Am J Cardiol 1997;79:10.

4. Frustaci A, Priori SG, Pieroni M, et al. Cardiac histological substrate inpatients with clinical phenotype of Brugada syndrome. Circulation2005;112:3680.

5. Meregalli PG, Tan HL, Probst V, et al. Type of SCN5A mutationdetermines clinical severity and degree of con duction slowing in loss-of-function sodium channelopathies. Heart Rhythm 2009;6:341.

6. Yan GX, Antzelevitch C. Cellular basis for the Brugada syndrome andother mechanisms of arrhythmogenesis associated with ST-segmentelevation. Circulation 1999;100:1660.

7. Postema PG, van Dessel PF, Kors JA, et al. Local depolarizationabnormalities are the dominant pathophysiologic mechanism for type 1electrocardiogram in Brugada syndrome: a study of electrocardiograms,vectorcardiograms, and body surface potential maps during ajmalineprovocation. J Am Coll Cardiol 2010;55:789.

8. LambiasePD,AhmedAK,CiaccioEJ, et al.High-density substratemappingin Brugada syndrome: combined role of conduction and repolarizationheterogeneities in arrhythmogenesis. Circulation 2009;120:106.

9. Nagase S, Kusano KF, Morita H, et al. Epicardial electrogram of theright ventricular outflow tract in patients with the Brugada syndrome:using the epicardial lead. J Am Coll Cardiol 2002;39:1992.

10. Nademanee K, Veerakul G, Chandanamattha P, et al. Prevention ofventricular fibrillation episodes in Brugada syndrome by catheterablation over the anterior right ventricular outflow tract epicardium.Circulation 2011;123:1270.

11. Tukkie R, Sogaard P, Vleugels J, et al. Delay in right ventricular activationcontributes to Brugada syndrome. Circulation 2004;109:1272.

12. BorggrefeM, Schimpf R. J-wave syndromes caused by repolarization ordepolarization mechanisms a debated issue among experimental andclinical electrophysiologists. J Am Coll Cardiol 2010;55:798.

13. Elizari MV, Levi R, Acunzo RS, et al. Abnormal expression ofcardiac neural crest cells in heart development: a different hypothesisfor the etiopathogenesis of Brugada syndrome. Heart Rhythm2007;4:359.

14. Peréz-Riera AR, Ferreira Filho C, de Abreu LC, ET al; On behalf of theInternational VCG Investigators Group. Do patients with electrocar-diographic Brugada type 1 pattern have associated right bundle branchblock? A comparative vectorcardiographic study. Europace. 2012 Jan10. [Epub ahead of print]

15. Hoogendijk MG, Potse M, Linnenbank AC, et al. Mechanism of rightprecordial ST-segment elevation in structural heart disease: excitationfailure by current-to-load mismatch. Heart Rhythm 2010;7:238.

16. Brugada P. Commentary on the Brugada ECG pattern. Circ ArrhythmElectrophysiol 2010;3:280.

17. Antzelevitch C, Brugada R. Fever and Brugada syndrome. Pacing ClinElectrophysiol 2002;25:1537.

18. Samani K, Wu G, Ai T, et al. A novel SCN5A mutation V1340I inBrugada syndrome augmenting arrhythmias during febrile illness. HeartRhythm 2009;6:1318.

19. Yap YG, Behr ER, Camm AJ. Drug-induced Brugada syndrome.Europace 2009;11:989.

20. Baranchuk A, Nguyen T, Hyung M et al. Brugada phenocopy: newterminology and proposed classification. Ann NonInvasive Electro-cardiol 2012 (in press).

21. Postema PG, Wolpert C, Amin AS, et al. Drugs and Brugada syndromepatients: Review of the literature, recommendations, and an up-to-dateWeb site (www.brugadadrugs.org). Heart Rhythm 2009;6:1335.

22. Garcia Niebla J, Llontop-Garcia P, Valle J, et al. Technical mistakesduring the acquisition of the electrocardiogram. Ann NoninvasiveElectrocardiol 2009;14:389.

23. Kligfield P, Gettes L, Bailey JJ, et al. Recommendations for thestandardization and the interpretation of the ECG: part 1: the ECG andits technology. J Am Coll Card 2008;49:1109.

24. Wilde AA, Antzelevitch C, Borggrefe M, et al, Study Group on theMolecular Basis of Arrhythmias of the European Society of Cardiology.Proposed diagnostic criteria for the Brugada syndrome: consensusreport. Circulation 2002;106:2514.

25. Antzelevitch C, Brugada P, Borggrefe M, et al. Brugada syndrome:report of the second consensus conference. Heart Rhythm 2005;2:429.

26. Chevallier S, Forclaz A, Tenkorang J, et al. New electrocardiographiccriteria for discriminating between Brugada types 2 and 3 patterns andincomplete right bundle branch block. J Am Coll Cardiol 2011;58:2290.

27. Ohkubo K, Watanabe I, Okumura Y, et al. A new criteria differentiatingtype 2 and 3 Brugada patterns from ordinary incomplete right bundlebranch block. Int Heart J 2011;52:159.

28. Nishizaki M, Sugi K, Izumida N, et al. Classification and assessment ofcomputerized diagnostic criteria for Brugada-type electrocardiograms.Heart Rhythm 2010;7:1660.

29. Bayés de Luna A, Garcia Niebla J, Goldwasser D. The ECGdifferential diagnosis of Brugada pattern and athletes. Eur Heart J2010 [May 14 (E-Public)].

30. Kurita T, Shimizu W, Inagaki M, et al. The electrophysiologicmechanism of ST segment elevation in Brugada syndrome. J Am CollCardiol 2002;40:330.

31. Serra G, Goldwasser D, Capulzini L, et al. New ECG criteria taken fromr' characteristics for differentiate type 2 Brugada pattern fromincomplete RBBB in athletes. Abstract accepted to European Congressof Cardiology; 2012.

32. Veltman C, Papavassilu T, Konrad M, et al. Insights into the location oftype 1 ECG in patients into the location of type 1 ECG in patients withBrugada syndrome: correlations of ECG and CV magnetic resonanceimaging. Heart Rhythm 2012;9:414.

33. Richter S, Sarkozy A, Veltmann C, et al. Variability of thediagnostic ECG pattern in an ICD patient population with Brugadasyndrome. J Cardiovasc Electrophysiol 2009;20:69.

34. Jayroe JB, Spodick DH, Nikus K, et al. Differentiating ST elevationmyocardial infarction and nonischemic causes of ST elevation byanalyzing the presenting electrocardiogram. Am J Cardiol 2009;103:301.

35. Oreto G, Corrado D, Delise P, et al. Doubts of the cardiologist regarding anelectrocardiogram presenting QRS V1-V2 complexes with positiveterminalwave and ST segment elevation. Consensus Conference promotedby the Italian Cardiology Society. G Ital Cardiol (Rome) 2010;11:3S.

36. Corrado D, Pelliccia A, Heidbuchel H, et al. Section of SportsCardiology, European Association of Cardiovascular Prevention andRehabilitation; Working Group of Myocardial and Pericardial Disease,European Society of Cardiology. Recommendations for interpretationof 12-lead electrocardiogram in the athlete. Eur Heart J 2010;31:243.

37. Karaoka H. Electrocardiographic patterns of the Brugada syndrome intwo young patients with pectus excavatum. J Electrocardiol2002;35:169.

38. Marcus FI, McKennaWJ, Sherrill D, et al. Diagnosis of arrhythmogenicright ventricular cardiomyopathy/dysplasia. Proposed modification ofthe Task Force Criteria. Circulation 2010;121:1533.

39. Yodogawa K, Morita N, Kobayashi Y, et al. A new approach for thecomparison of conduction abnormality between arrhythmogenic rightventricular cardiomyopathy/dysplasia and Brugada syndrome. AnnNoninvasive Electrocardiol 2011;16:263.

40. Haïssaguerre M, Derval N, Sacher F, et al. Sudden cardiac arrestassociated with early repolarization. N Eng J Med 2008;358:2016.

41. Rosso R, Kogan E, Belhassen B, et al. J-point elevation in survivors ofprimary ventricular fibrillation and matched control subjects: incidenceand clinical significance. J Am Coll Cardiol 2008;52:1231.

42. McIntyre W, Perez Riera A, Femenia F, et al. Coexisting earlyrepolarization pattern and Brugada syndromes recognition of potentiallyoverlapping entities. J Electrocardiol 2012;45:195.

43. Pitzalis MV, Anaclerio M, Iacoviello M. QT interval prolongation inright precordial leads in Brugada syndrome. J Am Coll Cardiol2003;42:1632.

442 A. Bayés de Luna et al. / Journal of Electrocardiology 45 (2012) 433–442

44. Babai Bigi M, Aslani A, Shahrzad S. aVR sign as a risk factor for life-threatening arrhythmic events in patients with Brugada syndrome. HeartRhythm 2007;4:1009.

45. Sarkozy A, Cherchia G, Paparella G, et al. Inferior and lateral ECGrepolarization abnormalities in Brugada syndrome. Circ ArrhythmicElectrophysiol 2009;2:154.

46. Morita H, Kusano KF, Miura D, et al. Fragmented QRS as a marker ofconduction abnormality and a predictor of prognosis of Brugadasyndrome. Circulation 2008;118:1697.

47. Tada T, Kusano KF, Nagase S. The relationship between the magnitudeof T wave alternans and the amplitude of the T wave in Brugadasyndrome. J Cardiovasc Electrophysiol 2008;19:56.

48. Makimoto H, Nakagawa E, Takaki H, et al. Augmented ST-segment elevation during recovery from exercise predicts cardiac

events in patients with Brugada syndrome. J Am Coll Cardiol2010;56:1576.

49. Huang Z, Patel C, Li W, et al. Role of signal-averaged electrocardio-grams in arrhythmic risk stratification of patients with Brugadasyndrome: a prospective study. Heart Rhythm 2009;8:1156.

50. Brugada P, Brugada R, Mont L, et al. Natural history of Brugadasyndrome: the prognostic value of programmed electrical stimulation ofthe heart. J Cardiovasc Electrophysiol 2003;14:455.

51. Priori SG, Napolitano C, Gasparini M, et al. Natural history of Brugadasyndrome: insights for risk stratification and management. Circulation2002;105:1342.

52. Probst V, Veltmann C, Eckardt L, et al. Long-term prognosis of patientsdiagnosed with Brugada syndrome: results from the FINGER BrugadaSyndrome Registry. Circulation 2010;121:635.