J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y VO L . 5 , N O . 4 , 2 0 1 9

ª 2 0 1 9 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

Comparison of Ajmaline andProcainamide Provocation Tests in theDiagnosis of Brugada Syndrome

Christopher C. Cheung, MD,a Greg Mellor, MD,b,c Marc W. Deyell, MD, MSC(EPI),a Bode Ensam, MBBS,b,c

Velislav Batchvarov, MD,b,c Michael Papadakis, MD,b,c Jason D. Roberts, MD, MAS,d Richard Leather, MD,e

Shubhayan Sanatani, MD,f Jeffrey S. Healey, MD,g Vijay S. Chauhan, MD,h David H. Birnie, MB, CHB,i

Jean Champagne, MD,j Paul Angaran, MD,k George J. Klein, MD,d Raymond Yee, MD,d Christopher S. Simpson, MD,l

Mario Talajic, MD,m Martin Gardner, MD,n John A. Yeung-Lai-Wah, MD,a Santabhanu Chakrabarti, MD,a

Zachary W. Laksman, MD, MSC,a Sanjay Sharma, MD,b,c Elijah R. Behr, MD,b,c Andrew D. Krahn, MDa

ABSTRACT

ISS

Fro

Ac

Cli

Ca

Co

Co

tut

OnkD

Un

Mo

OBJECTIVES The authors studied the response rates and relative sensitivity of the most common agents used in the

sodium-channel blocker (SCB) challenge.

BACKGROUND A type 1 Brugada electrocardiographic pattern precipitated by an SCB challenge confers a diagnosis of

Brugada syndrome.

METHODS Patients undergoing an SCB challenge were prospectively enrolled across Canada and the United Kingdom.

Patients with no prior cardiac arrest and family histories of sudden cardiac death or Brugada syndrome were included.

RESULTS Four hundred twenty-five subjects underwent SCB challenge (ajmaline, n ¼ 331 [78%]; procainamide, n ¼ 94

[22%]), with a mean age of 39 � 15 years (54% men). Baseline non–type 1 Brugada ST-segment elevation was present in

10%. A total of 154 patients (36%) underwent signal-averaged electrocardiography, with 41% having late potentials.

Positive results were seen more often with ajmaline than procainamide infusion (26% vs. 4%, p < 0.001). On multivariate

analysis, baseline non–type 1 Brugada ST-segment elevation (odds ratio [OR]: 6.92; 95% confidence interval [CI]: 3.15 to

15.2; p < 0.001) and ajmaline use (OR: 8.76; 95% CI: 2.62 to 29.2; p < 0.001) were independent predictors of positive

results to SCB challenge. In the subgroup undergoing signal-averaged electrocardiography, non–type 1 Brugada ST-

segment elevation (OR: 9.28; 95% CI: 2.22 to 38.8; p ¼ 0.002), late potentials on signal-averaged electrocardiography

(OR: 4.32; 95% CI: 1.50 to 12.5; p ¼ 0.007), and ajmaline use (OR: 12.0; 95% CI: 2.45 to 59.1; p ¼ 0.002) were strong

predictors of SCB outcome.

CONCLUSIONS The outcome of SCB challenge was significantly affected by the drug used, with ajmaline more likely to

provoke a type 1 Brugada electrocardiographic pattern compared with procainamide. Patients undergoing SCB challenge

may have contrasting results depending on the drug used, with potential clinical, psychosocial, and socioeconomic

implications. (J Am Coll Cardiol EP 2019;5:504–12) © 2019 by the American College of Cardiology Foundation.

N 2405-500X/$36.00 https://doi.org/10.1016/j.jacep.2019.01.026

m the aDivision of Cardiology, University of British Columbia, Vancouver, British Columbia, Canada; bCardiology Clinical

ademic Group, St. George’s University Hospitals NHS Foundation Trust, London, United Kingdom; cInstitute of Molecular and

nical Sciences, St. George’s University of London, London, United Kingdom; dSection of Cardiac Electrophysiology, Division of

rdiology, Western University, London, Ontario, Canada; eDivision of Cardiology, Royal Jubilee Hospital, Victoria, British

lumbia, Canada; fChildren’s Heart Centre, Department of Pediatrics, University of British Columbia, Vancouver, British

lumbia, Canada; gDivision of Cardiology, McMaster University, Hamilton, Ontario, Canada; hToronto General Research Insti-

e, University Health Network, Toronto, Ontario, Canada; iUniversity of Ottawa Heart Institute, University of Ottawa, Ottawa,

tario, Canada; jInstitut Universitaire de Cardiologie et Pneumologie de Québec, Université Laval, Québec City, Québec, Canada;

ivision of Cardiology, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada; lDivision of Cardiology, Queen’s

iversity, Kingston, Ontario, Canada; mInstitut de Cardiologie de Montréal, Département of Médecine, Université de Montréal,

ntréal, Québec, Canada; and the nDivision of Cardiology, Dalhousie University, Halifax, Nova Scotia, Canada. This study was

AB BR E V I A T I O N S

AND ACRONYM S

BrS = Brugada syndrome

ECG = electrocardiography

HLECG = high precordial lead

electrocardiography

RMS = root-mean-square

SADS = sudden arrhythmic

death syndrome

SAECG = signal-averaged

electrocardiography

SCB = sodium-channel blocker

SCD = sudden cardiac death

= ST-segment elevation

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9 Cheung et al.A P R I L 2 0 1 9 : 5 0 4 – 1 2 Sodium-Channel Blockers in Suspected Brugada Syndrome

505

B rugada syndrome (BrS) is diagnosed in pa-tients with a spontaneous type 1 Brugadaelectrocardiographic pattern, defined as

coved-type ST-segment elevation (STE) with $2 mmin $1 lead among the right precordial leads in stan-dard and high lead positions (i.e., V1 and V2

positioned in the fourth, third, or second intercostalspaces) (1). In patients with nondiagnostic baselineelectrocardiographic findings, a drug provocationchallenge using a sodium-channel blocker (SCB) maybe used to provoke a type 1 Brugada pattern (1). Inprevious guidelines, a provoked type 1 Brugadapattern unmasked through SCB challenge (i.e., drug-induced type 1) was considered diagnostic for BrS,and thus patients should receive appropriate riskstratification and management similar to patientswith a spontaneous type 1 Brugada pattern (1). How-ever, a recent consensus statement has placed lessemphasis on a drug-induced type 1 Brugada pattern,advocating the use of the proposed Shanghai score (2).

SEE PAGE 513

Although a spontaneous Brugada electrocardio-graphic pattern confers the greatest risk for suddencardiac death (SCD), a provoked Brugada pattern (i.e.,fever, drug provocation) is sufficient for diagnosisand may confer an increased risk for SCD as well (1).Furthermore, a diagnosis of BrS may lead to signifi-cant changes in clinical care and/or sequelae(including implantable cardioverter-defibrillatorplacement) and may also have psychosocial and so-cioeconomic impacts on those diagnosed (3). MultipleClass I antiarrhythmic drugs are used internationallyfor the purpose of SCB challenge (i.e., ajmaline, fle-cainide, pilsicainide, and procainamide), but therelative potencies of these drug for provoking a type 1Brugada electrocardiographic pattern are unclear.Previous small studies have compared intravenousflecainide and ajmaline, demonstrating greaterresponse to ajmaline (4,5). Further comparison ofagents is hampered by a lack of availability of allagents in any single country, and commonly only asingle intravenous form is available in most coun-tries. A differential response to SCB challenge mayaffect diagnostic rates and in turn affect the

supported by the Heart and Stroke Foundation of Canada (G-13-0002775 an

Robert Lancaster Memorial fund sponsored by McColl’s RG. The authors have

the contents of this paper to disclose.

All authors attest they are in compliance with human studies committees

stitutions and Food and Drug Administration guidelines, including patient co

the JACC: Clinical Electrophysiology author instructions page.

Manuscript received October 29, 2018; revised manuscript received January

subsequent management for patients withsuspected BrS and their families. Our objec-tive was to compare the relative yield ofajmaline and procainamide, the mostcommonly used agents in Europe and NorthAmerica, respectively, in provoking a type 1Brugada electrocardiographic pattern.

METHODS

PATIENT ENROLLMENT. Patients were iden-tified from several prospective registries.First, the CASPER (Cardiac Arrest SurvivorsWith Preserved Ejection Fraction) registryenrolls patients and families with histories of

sudden unexplained death or cardiac arrest fromacross Canada (6,7). Second, local inheritedarrhythmia registries from Vancouver, BritishColumbia, and London, Ontario, routinely enrollfamilies referred for suspected inherited arrhythmiasor BrS. In this combined Canadian cohort, the SCBchallenge (using procainamide) was conducted inkeeping with the CASPER diagnostic algorithm or atthe discretion of the local investigator on the basis ofthe index of suspicion from the context of event,family history, and related diagnoses (6). The secondcohort was a prospective institutional registry at St.George’s, University of London, composed of patientsundergoing SCB challenge (with ajmaline) as part of acomprehensive evaluation for a family history ofsudden arrhythmic death syndrome (SADS; suddenunexplained death with normal postmortem findings)as previously described (8). All subjects underwenttransthoracic echocardiography to exclude structuralheart disease. To account for differences betweencohorts, patients were matched for a positive familyhistory of SCD, SADS, or BrS but no personal history ofcardiac arrest. All patients were required to haveundergone 12-lead electrocardiography (ECG), highprecordial lead ECG (HLECG), and SCB challenge to beenrolled in this study. Patients with a spontaneoustype 1 Brugada electrocardiographic pattern at base-line were excluded. The research ethics board at eachinstitution approved the protocol.

DATA COLLECTION. Data collection consisted ofdemographics and clinical history and cardiac

STE

d G-14-0005732), Cardiac Risk in the Young, and the

reported that they have no relationships relevant to

and animal welfare regulations of the authors’ in-

nsent where appropriate. For more information, visit

2, 2019, accepted January 31, 2019.

FIGURE 1 Example of Positive Result on Sodium-Channel

Blocker Challenge

Sequential electrocardiograms (ECGs) (standard and high

precordial lead positions) during a sodium-channel blocker

challenge. The baseline ECG demonstrates a non–type 1 Brugada

ST-segment elevation, followed by the induction of a type 1

Brugada pattern during the drug challenge and resolution with

isoproterenol infusion.

Cheung et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9

Sodium-Channel Blockers in Suspected Brugada Syndrome A P R I L 2 0 1 9 : 5 0 4 – 1 2

506

investigations, including 12-lead ECG, HLECG (leadsV1 and V2 in the second and third intercostal spaces),signal-averaged ECG (SAECG), and SCB challenge (7).Results of baseline ECG and HLECG were classified asnormal or non–type 1 Brugada STE if either type 2 ortype 3 Brugada electrocardiographic pattern waspresent. Standard definitions of Brugada electrocar-diographic types 1 to 3 were used (9). The results ofSCB challenge were considered positive with theprecipitation of a type 1 Brugada electrocardiographicpattern in $1 lead in either standard or high leadpositions (Figure 1) (1). The results of SAECG wereconsidered abnormal if 1 or more positive parameterswere present (i.e., filtered QRS duration >114 ms, low-amplitude signal duration >38 ms, terminal QRS rootmean square voltage <20 mV).

SCB CHALLENGE. All Canadian sites used a standardSCB challenge protocol (7). Procainamide was infusedthrough a peripheral intravenous line with contin-uous electrocardiographic monitoring at a dose of 15mg/kg (maximum 1,000 mg) at 50 mg/min. In contrast

to previous studies in which a dose of 10 mg/kg wasadministered at 100 mg/min, the infusion protocolwas adapted to comply with the product monographin Canada, and by doing so, a higher total dose wasadministered at a slower rate, thereby enhancingsensitivity (7). Standard 12-lead ECG and HLECG wereperformed at baseline and were repeated at 10-minintervals during the infusion and at 30-min intervalsfor 1 h following completion of the infusion (CentralIllustration).

The United Kingdom site administered ajmalineintravenously at 1 mg/kg (maximum 100 mg) over 5min with continuous electrocardiographic recordingsfrom baseline until the electrocardiogram normal-ized, with simultaneous 15-lead electrocardiographicrecordings including high precordial leads as previ-ously described (10). For both protocols, the infusionwas terminated if a type 1 Brugada electrocardio-graphic pattern was provoked, the QRS durationincreased by $130% from baseline, premature ven-tricular complexes or ventricular arrhythmias devel-oped, or any significant side effects were noted.

STATISTICAL ANALYSIS. Statistical analysis wasperformed using StataIC version 15.0 (StataCorp,College Station, Texas). Comparisons were performedusing chi-square and univariate logistic regressionanalyses. Variables demonstrating significant associ-ation on univariate analysis (p < 0.05) were includedin a multivariate logistic regression. Patients whounderwent SAECG were included in a secondarysubgroup analysis. The dependent outcome for allunivariate and multivariate analyses was theresponse to SCB challenge.

RESULTS

BASELINE CLINICAL CHARACTERISTICS OF FULL

COHORT. Four hundred twenty-five patients (meanage 39 � 15 years, 54% men) were enrolled and un-derwent all pre-requisite investigations. Clinicalcharacteristics including age, sex, ethnicity, historyof syncope, indications for testing, and baselineelectrocardiographic findings are presented in Table 1.There were 42 patients (10%) with baseline non–type1 Brugada STE in standard (5%) or high (9%) pre-cordial leads. The procainamide and ajmaline groupswere of similar age (38.3 � 15.6 years vs. 39.4 � 14.8years; p ¼ 0.562), with the procainamide group hav-ing a higher proportion of non-Caucasian patients(52% vs. 6%; p < 0.001) and patients with baselinenon–type 1 Brugada STE (16% vs. 8%; p ¼ 0.031).

SAECG SUBGROUP. A total of 154 patients (36%) un-derwent SAECG as part of their assessment (Table 1).

CENTRAL ILLUSTRATION SCB Provocation for Brugada Syndrome

Cheung, C.C. et al. J Am Coll Cardiol EP. 2019;5(4):504–12.

CI ¼ confidence interval; ECG ¼ electrocardiogram; ICS ¼ intercostal space; OR ¼ odds ratio; SCB ¼ sodium channel blocker;

STE ¼ ST-segment elevation.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9 Cheung et al.A P R I L 2 0 1 9 : 5 0 4 – 1 2 Sodium-Channel Blockers in Suspected Brugada Syndrome

507

TABLE 1 Clinical Characteristics of the Combined Cohort

CombinedCohort

(n ¼ 425)

AjmalineCohort

(n ¼ 331)

ProcainamideCohort(n ¼ 94) p Value

Age, yrs 39.1 � 15.0 39.4 � 14.8 38.3 � 15.6 0.562

Male 231 (54) 185 (56) 46 (49) 0.243

Ethnicity

Caucasian 357 (84) 312 (94) 45 (48) <0.001

Asian 37 (9) 17 (5) 20 (21) <0.001

Other 31 (7) 2 (1) 29 (31) <0.001

Family history of Brugada, SCD or UCA 425 (100) 331 (100) 94 (100) 1.00

Brugada or IA syndrome 49 (12) 0 (0) 49 (52) <0.001

SCD or UCA 376 (88) 331 (100) 45 (48) <0.001

Syncope 29 (7) 21 (6) 8 (9) 0.488

Baseline ECG

Non–type 1 Brugada STE (standard leads) 20 (5) 8 (2) 12 (13) <0.001

Non–type 1 Brugada STE (high leads) 37 (9) 27 (8) 10 (11) 0.533

Non–type 1 Brugada STE (any leads) 42 (10) 27 (8) 15 (16) 0.031

SAECG 154 (36) 97 (29) 57 (61) <0.001

Filtered QRS duration, ms 112 � 13 110 � 10 115 � 17 0.028

Low-amplitude signal duration, ms 33.8 � 11.2 32.8 � 9.0 35.6 � 14.0 0.135

Terminal QRS RMS voltage, mV 35.5 � 19.8 36.6 � 19.6 33.7 � 20.1 0.377

Late potentials on SAECG

0 of 3 abnormal parameters 91 (59) 64 (66) 27 (47) 0.028

1 of 3 abnormal parameters 26 (17) 12 (12) 14 (25) 0.074

2 of 3 abnormal parameters 12 (8) 8 (8) 4 (7) 1.00

3 of 3 abnormal parameters 25 (16) 13 (13) 12 (21) 0.260

Positive result on SCB challenge

Combined cohort 89 (21) 85 (26) 4 (4) <0.001

SAECG subgroup 21 (14) 18 (19) 3 (5) 0.027

Values are mean � SD or n (%).

ECG ¼ electrocardiography; IA ¼ inherited arrhythmia; RMS ¼ root mean square; SAECG ¼ signal averagedelectrocardiography; SCB ¼ sodium-channel blocker; SCD ¼ sudden cardiac death; STE ¼ ST-segment elevation;UCA ¼ unexplained cardiac arrest.

TABLE 2 Factors Associated With Positive Results on Sodium-Channel Blocker Challenge

in the Combined Cohort (N ¼ 425)

SCB ChallengeResult Univariate Multivariate

Positive Negative OR (95% CI) p Value OR (95% CI) p Value

Age (yrs)

<26 16 (4) 78 (18) 1.00 (Ref.) NA

26–50 45 (11) 173 (41) 1.28 (0.68–2.41) 0.436

>50 28 (7) 84 (20) 1.65 (0.83–3.27) 0.155

Continuous NA NA 1.02 (1.00–1.03) 0.053

Male 47 (11) 184 (43) 0.92 (0.58–1.48) 0.742 1.02 (1.00–1.03) 0.067

Ethnicity

Caucasian 84 (20) 273 (64) 1.00 (Ref.) NA 1.00 (Ref.) n/a

Asian 4 (1) 33 (8) 0.39 (0.14–1.14) 0.087 0.78 (0.24–2.48) 0.668

Other orunspecified

1 (0) 30 (7) 0.11 (0.01–0.81) 0.030 0.39 (0.04–3.64) 0.408

Non–type 1Brugada STE

20 (5) 22 (5) 4.14 (2.14–8.00) <0.001 6.92 (3.15–15.2) <0.001

SCB

Procainamide 4 (1) 90 (21) 1.00 (Ref.) NA 1.00 (Ref.) NA

Ajmaline 85 (20) 246 (58) 7.77 (2.77–21.8) <0.001 8.76 (2.62–29.2) <0.001

Values are n (%) unless otherwise indicated.

CI ¼ confidence interval; NA ¼ not applicable; OR ¼ odds ratio; other abbreviations as in Table 1.

Cheung et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9

Sodium-Channel Blockers in Suspected Brugada Syndrome A P R I L 2 0 1 9 : 5 0 4 – 1 2

508

SAECG patients were similar with respect to baselineclinical characteristics (Online Table 1). Late poten-tials ($1 abnormal parameter) were present in 63 pa-tients in the combined cohort (41%). Thirty-sevenpatients (24%) had $2 abnormal parameters.Abnormal findings on SAECG were more frequent inthe procainamide group than the ajmaline group (53%vs. 34%; p ¼ 0.028).

SCB CHALLENGE. Four hundred twenty-five patientsunderwent the SCB challenge with either ajmaline (n¼ 331 [78%]) or procainamide (n ¼ 94 [22%]). In thecombined cohort, 89 patients (21%) had positive re-sults on SCB challenge, with ajmaline patients morelikely to have positive results compared with procai-namide patients (26% vs. 4%; p < 0.001). In theSAECG subgroup (n ¼ 154), 21 patients (14%) hadpositive results on SCB challenge, with ajmaline pa-tients again more likely to have positive results (19%vs. 5%; p ¼ 0.027).

PREDICTORS OF POSITIVE SCB CHALLENGE

RESULTS. Univariate and multivariate analyses wereperformed to identify predictors of positive results onSCB challenge (Table 2). In the combined cohort,increasing age, ethnicity, non–type 1 Brugada STE atbaseline, and ajmaline use were associated withpositive results on SCB challenge on univariate anal-ysis. On multivariate analysis, non–type 1 BrugadaSTE (odds ratio: 6.92; 95% confidence interval: 3.15 to15.2; p < 0.001) and ajmaline use (odds ratio: 8.76;95% confidence interval: 2.62 to 29.2; p < 0.001) wereindependently associated with positive results onSCB challenge (Figure 2, Central Illustration). Patientswith positive SCB results provoked by ajmaline orprocainamide were similar with respect to baselineclinical characteristics, but ajmaline-positive patientswere less likely to have non–type 1 Brugada STE atbaseline compared with procainamide-positive pa-tients (standard leads, p ¼ 0.002; any leads, p ¼0.034) (Online Table 2).

In the SAECG subgroup (n ¼ 154), non–type 1 Bru-gada STE, late potentials on SAECG, and ajmaline usewere associated with positive results on SCB chal-lenge on univariate analysis (Table 3). These variablesremained associated with positive challenge resultson multivariate analysis, with ajmaline use demon-strating the strongest association (odds ratio: 12.0;95% confidence interval: 2.45 to 59.1; p ¼ 0.002).

DISCUSSION

In a cohort of 425 subjects with family histories ofsudden death, SADS, or BrS who underwent SCBchallenge for the diagnosis of BrS, there was a marked

FIGURE 2 Factors Associated With Positive Results on Sodium-Channel

Blocker Challenge

Representative Figure (combined cohort). Forest plot of predictors of

sodium-channel blocker (SCB) challenge in combined cohort (n ¼ 425) and

signal-averaged electrocardiography (SAECG) subgroup (n ¼ 154) on multi-

variate analysis. In the combined cohort, independent predictors of positive

results on sodium-channel blocker challenge included non–type 1 Brugada

ST-segment elevation (STE) (odds ratio [OR]: 6.92) and ajmaline use (OR:

8.76). In the SAECG subgroup, independent predictors included non–type 1

Brugada STE (OR: 9.28), late potentials on SAECG (OR: 4.32), and ajmaline use

(OR: 12.0).

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9 Cheung et al.A P R I L 2 0 1 9 : 5 0 4 – 1 2 Sodium-Channel Blockers in Suspected Brugada Syndrome

509

difference in the likelihood of a positive result withthe use of ajmaline compared with procainamide.After adjustment for other predictors, including malesex, ethnicity, familial clustering, baseline STE, andlate potentials on SAECG, ajmaline provocation wasthe strongest predictor of positive results on SCBchallenge. These results suggest that ajmaline issignificantly more potent than procainamide forprecipitating a type 1 electrocardiographic patternand raise concern about diagnostic accuracy. This isfurther compounded by the lack of a clear gold stan-dard for the diagnosis of BrS, limiting interpretationof sensitivity or specificity for either drug. It is notpossible to decide whether ajmaline has morefalse-positive results or if procainamide has morefalse-negative results. To our knowledge, this is thelargest study comparing 2 SCBs for the diagnosis ofBrS, using a matched cohort across 2 countries.

A more “sensitive” test may be seen as favorable,as more subjects at potential risk from sudden deathwill be identified, who had preventive measures maybe instituted. Indeed, sudden death victims with fa-milial BrS may not show a spontaneous type 1 elec-trocardiographic pattern prior to death, and cardiacarrest survivors with BrS may not show a sponta-neous pattern without provocation (7). However, ahigher “sensitivity” test may also lead to over-diagnosis, particularly when patients do not havebaseline STE (11). A diagnosis of BrS leads to signifi-cant clinical and psychological sequelae, which canoften be complicated by imperfect risk stratificationand a lack of effective medical therapies (apart fromquinidine in select high-risk patients) (12). As such, inthe absence of symptoms, patients with a drug-induced electrocardiographic pattern will not havesufficiently elevated risk for SCD to warrant specifictherapy, and “specificity” may be preferred to“sensitivity” (11). Importantly, these findings do notidentify the correct drug of choice in the assessmentof BrS, only that there is a significant difference be-tween procainamide and ajmaline. Determining thedrug of choice will likely depend on circumstance anda correlation of outcomes such as major arrhythmicevents.

Ajmaline is a potent inhibitor of the cardiac sodiumchannel and may be associated with precipitation of atype 1 pattern in the context of conditions other thanBrS. In a study of patients with atrioventricular nodalre-entrant tachycardia, 27% of patients and 4.5% ofhealthy control subjects had a type 1 electrocardio-graphic pattern when treated with ajmaline (13). Ge-netic testing revealed mutations or rare variants in upto 77%, although a large proportion of variantswere subsequently classified as benign or

nondamaging (13). The investigators hypothesizedthat loss-of-function sodium-channel mutations maysimultaneously lead to BrS and atrioventricular nodalre-entrant tachycardia through a preferential block of1 of the atrioventricular nodal pathways, althoughthis remains unproven (11). Peters et al. (14) alsoidentified positive ajmaline test results in 16% ofpatients with diagnoses of arrhythmogenic rightventricular cardiomyopathy. In a large study of familymembers with documented SCN5A mutations, ajma-line provocation was positive in 5.6% of genotype-negative subjects (i.e., false positives) (15). In arecent study of 637 patients evaluated for unex-plained cardiac arrest or SCD, 8% of families hadpositive ajmaline responses that were identified to bea confounder in the context of an alternative geneticdiagnosis or noncosegregation of the ajmalineresponse and arrhythmia (16). Conversely, we haveshown that 28% of systematically evaluated familiesof autopsy-negative SCD (SADS) victims yield a

TABLE 3 Factors Associated With Positive Results on Sodium-Channel Blocker Challenge

in the Signal-Averaged Electrocardiography Cohort (N ¼ 154)

SCB ChallengeResult Univariate Multivariate

Positive Negative OR (95% CI) p Value OR (95% CI) p Value

Age, yrs

<26 7 (5) 36 (23) 1.00 (Ref.)

26–50 11 (7) 66 (42) 0.86 (0.31–2.40) 0.769

>50 3 (2) 31 (20) 0.50 (0.12–2.09) 0.341

Continuous n/a n/a 1.00 (0.97–1.03) 0.761

Male 13 (8) 66 (43) 1.65 (0.64–4.24) 0.299

Ethnicity

Caucasian 18 (12) 102 (66) 1.00 (Ref.) NA

Asian 2 (1) 9 (6) 1.26 (0.25–6.31) 0.779

Other orunspecified

1 (1) 22 (14) 0.26 (0.03–2.03) 0.198

Non–type 1Brugada STE

6 (4) 14 (9) 3.40 (1.14–10.2) 0.029 9.28 (2.22–38.8) 0.002

SAECG

$1 abnormalparameter

13 (8) 50 (32) 2.70 (1.05–6.96) 0.040 4.32 (1.50–12.5) 0.007

$2 abnormalparameters

8 (5) 29 (19) 2.21 (0.83–5.83) 0.111

3 abnormalparameters

6 (4) 19 (12) 2.40 (0.83–6.96) 0.107

Per late potential NA NA 1.46 (1.01–2.12) 0.044

Sodium-channel blocker

Procainamide 3 (2) 54 (35) 1.00 (Ref.) NA 1.00 (Ref.) NA

Ajmaline 18 (12) 79 (51) 4.10 (1.15–14.6) 0.029 12.0 (2.45–59.1) 0.002

Values are n (%) unless otherwise indicated.

Abbreviations as in Tables 1 and 2.

Cheung et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9

Sodium-Channel Blockers in Suspected Brugada Syndrome A P R I L 2 0 1 9 : 5 0 4 – 1 2

510

positive ajmaline response (17). In another study ofSCB agents administered to 672 relatives with clearfamilial BrS, Therasse et al. (18) reported 54% asajmaline positive compared with 37% with flecainide,suggesting a tendency toward overdiagnosis withajmaline and underdiagnosis with flecainide.

Current guidelines state that BrS is diagnosed inthe presence of a type 1 ECG pattern either sponta-neously or after positive SCB provocation regardlessof symptomatic status (1). In the recent J-Wave Syn-dromes Expert Consensus Conference Report, theinvestigators recommended removing the label of BrSamong patients whose only type 1 Brugada patternwas drug induced, unless there is a symptom orfamily history that accompanies the finding, largelyin response to concerns about a poor “specificity” ofajmaline (2). In the proposed “Shanghai score,” adrug-induced type 1 electrocardiographic pattern isassigned fewer points than a spontaneous or fever-induced Brugada pattern (2). The results of ourstudy suggest further complexity in the interpreta-tion of a drug-induced ECG pattern, with the drugchoice significantly affecting “sensitivity.” Arguablyclinical context should primarily inform the use ofany test, and the context of a SADS death in an

immediate relative is important for the a priori like-lihood of BrS being present.

In the group in which SAECG was performed, latepotentials were also independently associated withpositive results on SCB challenge. This is in keepingwith recent evidence suggestive of conduction ab-normalities and myocardial fibrosis in BrS (19).Furthermore, a direct correlation between abnor-malities on SAECG and positive results on SCB chal-lenge has been suggested (20). Cumulatively, thepresence of late potentials and baseline STE may beuseful in counseling patients regarding undergoingSCB testing, but their absence does not eliminate theneed for SCB provocation when clinical suspicionremains.STUDY LIMITATIONS. There were differences in theclinical characteristics of the ajmaline and procai-namide groups, reflecting demographic differencesin the practices of the 2 recruiting groups.Although we matched patients for family history ofBrS or SCD, there remained significant differencesin the clinical characteristics between the 2 groups,with the possibility of residual confounding despiteour multivariate analysis. Notably, there were sig-nificant differences in baseline ethnicities betweenthe ajmaline and procainamide cohort, which wereincluded in our multivariate analysis. Withoutdirect comparison of the effect of each drug in thesame patient and clinical follow-up to identifydifferences in clinical events between the groups,it is not possible to recommend 1 drug over theother. Similarly, in the absence of a gold standard,it is impossible to determine whether the type 1Brugada pattern reflects a drug-induced and/ordose-dependent effect of the SCB and whetherajmaline has more false-positive results or if pro-cainamide has more false-negative results. Futurestudies should perform head-to-head comparisonsand include long-term follow-up to fully under-stand the clinical implications of an ajmalineversus procainamide-provoked type 1 Brugadapattern. Unfortunately, this was not possible,because of the lack of universal access to SCBs inCanada and the United Kingdom. In our cohort,genetic testing was performed on the basis ofclinician access and discretion in a small propor-tion of patients after the SCB challenge. Given thelow rate of pathogenic mutations and retrospectiveinterpretation (SCB challenge generally before ge-netic testing), we did not evaluate genetic resultsor incorporate this into our multivariate analysis.With a relatively heterogenous population of fam-ily members with no prior cardiac arrest, wealso did not report outcomes, because of the low

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE: Multiple Class I

antiarrhythmic drugs are used internationally in the SCB chal-

lenge (i.e., ajmaline, flecainide, pilsicainide, and procainamide),

but the relative potencies of each drug in provoking a type 1

Brugada electrocardiographic pattern are unclear. Although a

spontaneous type 1 Brugada electrocardiographic pattern con-

fers the greatest risk for SCD, a provoked type 1 Brugada pattern

is sufficient for diagnosis and may confer an increased risk for

SCD as well. A diagnosis of BrS may lead to significant changes in

clinical care and/or sequelae, including implanTable cardioverter-

defibrillator placement.

TRANSLATIONAL OUTLOOK 1: After adjustment for factors

associated with positive SCB challenge, the outcome of the SCB

challenge was significantly affected by the drug used.

TRANSLATIONAL OUTLOOK 2: Ajmaline was much more

likely to provoke a type 1 Brugada pattern, compared with pro-

cainamide. It is not possible to determine if ajmaline has more

false-positive results or if procainamide has more false-negative

results.

TRANSLATIONAL OUTLOOK 3: Varying sensitivities and

specificities depending on the agent used in the SCB challenge

may have potential clinical, psychosocial, and socioeconomic

implications.

J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9 Cheung et al.A P R I L 2 0 1 9 : 5 0 4 – 1 2 Sodium-Channel Blockers in Suspected Brugada Syndrome

511

rate of clinical events in follow-up (insufficientnumbers for comparison).

CONCLUSIONS

The outcome of the SCB provocation test for BrS wassignificantly affected by the agent used, with ajma-line substantially more likely to provoke a type 1Brugada electrocardiographic pattern than procaina-mide. These results have significant implications forthe “sensitivity” and “specificity” of the test in pa-tients suspected of BrS, albeit in the absence of a goldstandard. Ajmaline provocation is a more potent testthan procainamide provocation, raising concernsabout overdiagnosis and underdiagnosis of BrS,respectively. Large cohort head-to-head comparisonsare needed to identify the optimal testing strategy insuspected BrS.

ACKNOWLEDGMENTS The authors thank the Heartand Stroke Foundation of Canada, the UK CharityCardiac Risk in the Young, and McColl’s RG for theirgenerous funding and continued support. The au-thors thank Dr. Kathy Li for her statistical support inpreparing this paper.

ADDRESS FOR CORRESPONDENCE: Dr. AndrewKrahn, Heart Rhythm Vancouver, 211-1033 DavieStreet, Vancouver, British Columbia V6E 1M7, Can-ada. E-mail: akrahn@mail.ubc.ca.

RE F E RENCE S

1. Priori SG, Wilde AA, Horie M, et al.Executive summary: HRS/EHRA/APHRS expertconsensus statement on the diagnosis andmanagement of patients with inherited primaryarrhythmia syndromes. Europace 2013;15:1389–406.

2. Antzelevitch C, Yan GX, Ackerman MJ, et al. J-Wave syndromes expert consensus conferencereport: emerging concepts and gaps in knowledge.J Arrhythm 2016;32:315–39.

3. Mohammed S, Lim Z, Dean PH, et al. Geneticinsurance discrimination in sudden arrhythmiadeath syndromes: empirical evidence from a cross-sectional survey in North America. Circ CardiovasGenet 2017;10:e001442.

4. Wolpert C, Echternach C, Veltmann C, et al.Intravenous drug challenge using flecainide andajmaline in patients with Brugada syndrome. HeartRhythm 2005;2:254–60.

5. Conte G, Sieira J, Sarkozy A, et al. Life-threat-ening ventricular arrhythmias during ajmalinechallenge in patients with Brugada syndrome:

incidence, clinical features, and prognosis. HeartRhythm 2013;10:1869–74.

6. Krahn AD, Healey JS, Chauhan V, et al. Sys-tematic assessment of patients with unexplainedcardiac arrest: Cardiac Arrest Survivors With Pre-served Ejection Fraction Registry (CASPER). Cir-culation 2009;120:278–85.

7. Somani R, Krahn AD, Healey JS, et al. Procai-namide infusion in the evaluation of unexplainedcardiac arrest: from the Cardiac Arrest Survivorswith Preserved Ejection Fraction Registry (CAS-PER). Heart Rhythm 2014;11:1047–54.

8. Behr ER, Dalageorgou C, Christiansen M, et al.Sudden arrhythmic death syndrome: familialevaluation identifies inheritable heart disease inthe majority of families. Eur Heart J 2008;29:1670–80.

9. Antzelevitch C, Brugada P, Borggrefe M, et al.Brugada syndrome: report of the secondconsensus conference: endorsed by the HeartRhythm Society and the European Heart RhythmAssociation. Circulation 2005;111:659–70.

10. Govindan M, Batchvarov VN, Raju H, et al.Utility of high and standard right precordial leadsduring ajmaline testing for the diagnosis of Bru-gada syndrome. Heart 2010;96:1904–8.

11. Viskin S, Rosso R, Friedensohn L, Havakuk O,Wilde AA. Everybody has Brugada syndrome untilproven otherwise? Heart Rhythm 2015;12:1595–8.

12. Belhassen B, Glick A, Viskin S. Efficacy ofquinidine in high-risk patients with Brugada syn-drome. Circulation 2004;110:1731–7.

13. Hasdemir C, Payzin S, Kocabas U, et al. Highprevalence of concealed Brugada syndrome inpatients with atrioventricular nodal reentranttachycardia. Heart Rhythm 2015;12:1584–94.

14. Peters S, Trummel M, Denecke S, Koehler B.Results of ajmaline testing in patients witharrhythmogenic right ventricular dysplasia-car-diomyopathy. Int J Cardiol 2004;95:207–10.

15. Hong K, Brugada J, Oliva A, et al. Value ofelectrocardiographic parameters and ajmaline testin the diagnosis of Brugada syndrome caused bySCN5A mutations. Circulation 2004;110:3023–7.

Cheung et al. J A C C : C L I N I C A L E L E C T R O P H Y S I O L O G Y V O L . 5 , N O . 4 , 2 0 1 9

Sodium-Channel Blockers in Suspected Brugada Syndrome A P R I L 2 0 1 9 : 5 0 4 – 1 2

512

16. Tadros R, Nannenberg EA, Lieve KV, et al.Yield and pitfalls of ajmaline testing in the eval-uation of unexplained cardiac arrest and suddenunexplained death: single-center experience with482 families. J Am Coll Cardiol EP 2017;3:1400–8.

17. Papadakis M, Papatheodorou E, Mellor G, et al.The diagnostic yield of Brugada syndrome aftersudden death with normal autopsy. J Am CollCardiol 2018;71:1204–14.

18. Therasse D, Sacher F, Petit B, et al.Sodium-channel blocker challenge in the familial

screening of Brugada syndrome: safety and pre-dictors of positivity. Heart Rhythm 2017;14:1442–8.

19. Nademanee K, Raju H, de Noronha SV, et al.Fibrosis, connexin-43, and conduction abnor-malities in the Brugada syndrome. J Am CollCardiol 2015;66:1976–86.

20. Duthoit G, Fressart V, Hidden-Lucet F, et al.Brugada ECG pattern: a physiopathologicalprospective study based on clinical,

electrophysiological, angiographic, and geneticfindings. Front Physiol 2012;3:474.

KEY WORDS arrhythmia, Brugadasyndrome, cardiac arrest, sodium channels,sudden cardiac death

APPENDIX For supplemental tables, pleasesee the online version of this paper.