JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO.

7, 2022

ª 2022 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

PUBLISHED BY ELSEVIER

ORIGINAL RESEARCH

VENTRICULAR ARRHYTHMIAS - ELECTROCARDIOGRAPHY

Simplified Integrated Clinical and

Electrocardiographic Algorithm for
Differentiation of Wide QRS Complex
Tachycardia
The Basel Algorithm
Federico Moccetti, MD,a,b,c Mrinal Yadava, MD,b Yllka Latifi, MD,b,d Ivo Strebel, PHD,a Nikola Pavlovic, MD, PHD,a,e
Sven Knecht, DSC,a Babken Asatryan, MD, PHD,f Beat Schaer, MD,a Michael Kühne, MD,a Charles A. Henrikson, MD,b
Frank-Peter Stephan, MD,a Stefan Osswald, MD,a Christian Sticherling, MD,a Tobias Reichlin, MDa,f

ABSTRACT

BACKGROUND Prompt differential diagnosis of wide QRS complex tachycardia (WCT) is crucial to patient manage-
ment. However, distinguishing ventricular tachycardia (VT) from supraventricular tachycardia (SVT) with wide QRS
complexes remains problematic, especially for nonelectrophysiologists.

OBJECTIVES This study aimed to develop a simple-to-use algorithm with integration of clinical and electrocardio-
graphic (ECG) parameters for the differential diagnosis of WCT.

METHODS The 12-lead ECGs of 206 monomorphic WCTs (153 VT, 53 SVT) with electrophysiology-confirmed diagnoses
were analyzed. In the novel Basel algorithm, VT was diagnosed in the presence of at least 2 of the following criteria:
1) clinical high risk features; 2) lead II time to first peak >40 ms; and 3) lead aVR time to first peak >40 ms. The algorithm
was externally validated in 203 consecutive WCT cases (151 VT, 52 SVT). Its’ diagnostic performance and clinical
applicability were compared with those of the Brugada and Vereckei algorithms.

RESULTS The Basel algorithm showed a sensitivity, specificity, and accuracy of 92%, 89%, and 91%, respectively, in
the derivation cohort and 93%, 90%, and 93%, respectively, in the validation cohort. There were no significant
differences in the performance characteristics between the 3 algorithms. The evaluation of the clinical applicability of
the Basel algorithm showed similar diagnostic accuracy compared with the Brugada algorithm (80% vs 81%; P ¼ 1.00),
but superiority compared with the Vereckei algorithm (72%; P ¼ 0.03). The Basel algorithm, however, enabled a faster
diagnosis (median 36 seconds vs 105 seconds for the Brugada algorithm [P ¼ 0.002] and 50 seconds for the Vereckei
algorithm [P ¼ 0.02]).

CONCLUSIONS The novel Basel algorithm based on simple clinical and ECG criteria allows for a rapid and accurate
differential diagnosis of WCT. (J Am Coll Cardiol EP 2022;8:831–839) © 2022 by the American College of Cardiology
Foundation.

From the aDivision of Cardiology, Department of Medicine, University Hospital Basel, University of Basel, Basel, Switzerland;
b
Knight Cardiovascular Institute, Oregon Health and Science University, Portland, Oregon, USA; cHeart Center Lucerne, Luzerner
Kantonsspital, Lucerne, Switzerland; dCardiology Spital Uster, Uster, Switzerland; eCardiology, Klinicki Bolnicki Centar Sestre
Milosrdnice, Zagreb, Croatia; and the fDepartment of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern,
Switzerland.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’
institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information,
visit the Author Center.

Manuscript received October 18, 2021; revised manuscript received March 22, 2022, accepted March 28, 2022.

ISSN 2405-500X/$36.00 https://doi.org/10.1016/j.jacep.2022.03.017

832 Moccetti et al JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022

Basel Algorithm for Diagnosis of SVT/VT JULY 2022:831–839

A
ABBREVIATIONS wide QRS complex tachycardia patients undergoing EP studies for regular WCT at the
AND ACRONYMS (WCT) on the surface electrocardio- Knight Cardiovascular Institute of the Oregon Health
gram (ECG) prompts a complex dif- and Science University, Portland, Oregon, were
ECG = electrocardiography
ferential diagnosis of arrhythmias with retrospectively identified.
EP = electrophysiology
prognoses ranging from utterly benign to The study was performed in accordance with the
SVT = supraventricular
potentially lethal, requiring different treat- principles of the Declaration of Helsinki. The study
tachycardia
ment strategies.1,2 While WCT is a manifesta- protocol was approved by the local ethics
VT = ventricular tachycardia
tion of ventricular tachycardia (VT) in around committees.
WCT = wide QRS complex
80% of cases,2 other potential causes of WCT
tachycardia ECG ANALYSIS. The 12-lead ECGs were analyzed by 2
include pre-excited supraventricular tachy-
of the authors (F.M. and M.Y.) blinded to the EP
cardia (SVT), SVT with abnormal intraventricular con-
diagnosis at a sweep speed of 25 mm/s. As in previous
duction, ventricular-paced rhythm, and drug- and
studies assessing the accuracy of ECG algorithms, the
electrolyte-induced QRS widening.3,4 Rapid recogni-
12-lead ECGs of the tachycardias acquired during the
tion of the underlying cause of the WCT is of critical
EP studies were assessed.10,12 The criterion standard
importance for timely initiation of the appropriate
diagnosis of VT vs SVT was the one obtained during
treatment, 5 because delayed diagnosis or inappro-
the EP study. All ECGs from the derivation and vali-
priate management in patients with VT may have
dation cohorts were assessed according to: 1) the
deleterious consequences. 6-8 Therefore, a simplified,
novel simplified integrated ECG algorithm (“the Basel
algorithmic, and practical approach to WCTs is crucial
algorithm”); 2) the Brugada algorithm; 10 and 3) the
for all clinicians responsible for ECG interpretation,
Vereckei algorithm. 12 Interobserver agreement for the
whether in emergency medicine, cardiology, or pri-
measurements in lead II and aVR were assessed in the
mary care.9
validation cohort.
SEE PAGE 840
DERIVATION OF THE BASEL ALGORITHM. The per-
Over the past 3 decades, several criteria and algo- formance of several ECG candidate criteria was tested
rithms have been proposed to distinguish VT from in the derivation cohort and the 2 with the highest
SVT with abnormal intraventricular conduction.10-17 area under the receiver operating characteristic (ROC)
Among them the Brugada algorithm and Vereckei curve for the diagnosis of VT (lead II time to first
algorithm are the most commonly used. Despite peak, lead aVR time to first peak) were chosen.
the high sensitivity (SN) and specificity (SP) Accordingly, the proposed Basel algorithm is based on
reported in original cohorts, their use in the clinical 3 criteria (Figure 1, Central Illustration): 1) clinical high
setting has consistently showed low reproducibility risk feature; 2) lead II time to first peak >40 ms; and 3)
(Supplemental Table 1).18-23 Furthermore, these WCT lead aVR time to first peak >40 ms. VT was diagnosed
algorithms are rather complex, involving multiple if at least 2 of those 3 criteria were met. If 0 or 1 cri-
sequential steps and requiring evaluation of WCT terion was positive, the diagnosis was SVT. Examples
morphology, and are difficult to recall in an emer- of ECGs with WCT demonstrating successful and un-
gency setting. These factors render the everyday use successful diagnoses of VT and of SVT are shown in
of WCT algorithms challenging, particularly for non- Figure 2.
electrophysiologist (non-EP) practitioners. The clinical high risk feature criterion is considered
To address these remaining challenges in the dif- to be met if the patient has a history of myocardial
ferential diagnosis of WCT we aimed to develop a infarction, a history of congestive heart failure with
user-friendly algorithm based on clinical and ECG left ventricular ejection fraction #35% (or an
parameters to facilitate a rapid and accurate diagnosis implanted implantable cardioverter-defibrillator or
in patients presenting with monomorphic regular cardiac resynchronization therapy–defibrillator). The
WCT. lead II and lead aVR time to first peak criterion is
considered to be positive if the time from the begin-
ning of the QRS to the first change in polarity, ie, the
METHODS
first positive or negative deflection (or “peak”), is
>40 ms. This can be expressed as a QRS complex
PATIENT POPULATIONS AND STUDY DESIGN. For
beginning with an r- or R-wave >40 ms, or beginning
the derivation cohort, consecutive patients undergo-
with a q-, Q-, or QS-wave >40 ms.
ing electrophysiological (EP) studies for regular WCT
at the University Hospital of Basel, Switzerland, from EVALUATION OF THE CLINICAL APPLICABILITY OF
January 2010 to December 2014 were retrospectively THE NOVEL BASEL ALGORITHM. To evaluate the
identified. For the validation cohort, consecutive “real-world” clinical applicability of the novel
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022 Moccetti et al 833
JULY 2022:831–839 Basel Algorithm for Diagnosis of SVT/VT

F I G U R E 1 Graphical Representation of the Novel Basel Algorithm

CHF ¼ congestive heart failure; CRT ¼ cardiac resynchronization therapy; ICD ¼ implantable cardioverter-defibrillator; LVEF ¼ left ventricular
ejection fraction; SVT¼ supraventricular tachycardia; VT¼ ventricular tachycardia.

algorithm, a total of 8 physicians (2 EPs, 2 general DATA AVAILABILITY. The data underlying this
cardiologists, 2 cardiology fellows, and 2 internal article cannot be shared publicly owing to privacy of
medicine residents) analyzed a random subset of 50 the individuals that were investigated in the study.
WCT ECGs (25 VTs and 25 SVTs) at a sweep speed of The data will be shared on reasonable request to the
25 mm/s. Each of them analyzed the 50 ECGs indi- corresponding author provided that it in accordance
vidually, blinded to the criterion standard diagnosis. with the institutional ethical guidelines as well as
The ECG analysis was performed at 6 different time regulations and legislation.
points (with at least 2 weeks interval in between)
according to 5 existing algorithms (Brugada, Vereckei, RESULTS
Pava, Jastrzebski, and Chen) 10,13-16 and the novel
Basel algorithm. The time to diagnosis was recorded BASELINE CHARACTERISTICS OF THE PATIENTS. In
for every ECG assessed. the derivation cohort, a total of 206 WCT episodes
(153 VT, 53 SVT) were recorded in 124 patients. In the
STATISTICAL ANALYSIS. Continuous variables are
validation cohort, 203 WCT episodes (151 VT, 52 SVT)
presented as median (IQR) and compared with the use
were recorded in 112 patients. Baseline characteristics
of the Mann-Whitney U test. Chi-square tests were
of the patients and the ECG characteristics of the 2
used to compare categoric variables. To assess the
cohorts are presented in Tables 1 and 2. The under-
algorithm performance, SN, SP, positive predictive
lying arrhythmias in the groups of SVTs and VTs are
value, negative predictive value, and diagnostic ac-
presented in Table 3.
curacy were calculated from 2  2 cross-tables.
McNemar’s test was used to compare the perfor- DERIVATION OF THE NOVEL ALGORITHM. After
mance of the Basel algorithm with those of the Bru- analysis of several candidate criteria in the derivation
gada and Vereckei algorithms. Interobserver cohort, time to first peak in lead II and lead aVR
agreement for measurements performed in lead II showed the best performance characteristics (area
and aVR were assessed with the use of correlation under the ROC curve: 0.91 for both). ROC derived
coefficients and Bland-Altman plots in the validation optimal cutoffs were 51 ms for lead II time to peak and
cohort. 46 ms for lead aVR time to peak (Supplemental
For the clinical validation, SN, SP, positive pre- Table 2). For the algorithm, a cutoff of 40 ms was
dictive value, negative predictive value, diagnostic chosen to facilitate user-friendly application in clin-
accuracy, and average time to diagnosis using the 3 ical practice.
algorithms were compared by means of Tukey’s Observer agreement (Supplemental Figure 1)
multiple comparisons test. showed good correlation for lead II and aVR (r 2 ¼ 0.94
A P value <0.05 was considered to be statistically and r2 ¼0.92, respectively). Bland-Altman plots
significant. The statistical analyses were performed showed a small bias (0.4 ms and 0.8 ms for lead II
with the use of SPSS version 24.0 (IBM), Prism and aVR, respectively) and narrow 95% limits of
Graphpad version 8.2.1 (Graphpad Software), and R (R agreement (14.8 and 14.0 ms for lead II, 17.0 and
Foundation for Statistical Computing. 15.4 ms for lead aVR).
834 Moccetti et al JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022

Basel Algorithm for Diagnosis of SVT/VT JULY 2022:831–839

C E NT R AL IL L U STR AT IO N Simplified Integrated Clinical and Electrocardiographic Algorithm for Differentiation

of Wide QRS Complex Tachycardia

Derivation, 206 ECGs Novel Simplified Algorithm for the Differential Diagnosis of WCT
Validation, 203 ECGs
Lead II Lead aVR
I V1 Structural Heart
+ Time to First Peak + Time to First Peak
Disease
>40 ms >40 ms
Structural Heart Disease:
- Myocardial Infarction (history)
- CHF (LVEF <35%)
II V2 - Device (ICD, CRT)
120 ms

≥2 criteria fulfilled → VT 0 or 1 criteria fulfilled → SVT

III V3

Comparison of Algorithm Performance

100%
aVR V4
90%

80%

70%
aVL V5
60%

50%
Sensitivity Specificity Accuracy

aVF V6 Basel-Algorithm Derivation / Validation

Brugada-Algorithm Derivation / Validation
Vereckei-Algorithm Derivation / Validation

Clinical Validation

Diagnostic Accuracy Time to Diagnosis

P = 1.0 P = 0.002

P = 0.03 P = 0.02 P = 0.003 P = 0.02

100 160
90 140
80 120
Seconds

100
70
%

80
60
60
50 40
40 20
30 0
Brugada Vereckei Novel Brugada Vereckei Novel

EP Attending Cardiology Attending

Cardiology Fellow Internal Medicine Resident

Moccetti F, et al. J Am Coll Cardiol EP. 2022;8(7):831–839.

CHF ¼ congestive heart failure; CRT ¼ cardiac resynchronization therapy; ECG ¼ electrocardiogram; EP ¼ electrophysiology; ICD ¼ implantable cardioverter-
defibrillator; LVEF ¼ left ventricular ejection fraction; SVT ¼ supraventricular tachycardia; VT ¼ ventricular tachycardia; WCT ¼ wide QRS complex tachycardia.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022 Moccetti et al 835
JULY 2022:831–839 Basel Algorithm for Diagnosis of SVT/VT

F I G U R E 2 Examples of Electrocardiograms With Wide QRS Complex Tachycardia Demonstrating Successful and Unsuccessful Diagnoses
of VT and SVT

(A) Ventricular tachycardia (VT) with a focus in the left ventricular outflow tract correctly classified as VT with the Basel algorithm (time to first
peak >40 ms in leads II and aVR). (B) Fascicular VT incorrectly classified with the Basel algorithm (time to first peak <40 ms in leads II and
aVR). (C) Typical atrial flutter with 1:1 atrioventricular conduction and right bundle branch block aberrancy correctly classified as supraven-
tricular tachycardia (SVT) with the Basel algorithm (time to first peak <40 ms in leads II and aVR). (D) SVT incorrectly classified as VT with the
Basel algorithm (time to first peak >40 ms in leads II and aVR).

PERFORMANCE OF THE NOVEL ALGORITHM IN THE performance of each criterion, are presented in
DERIVATION AND THE VALIDATION COHORT. The Table 4. The novel algorithm reached SN and SP of,
diagnostic performance of the Brugada algorithm, the respectively, 91.5% and 88.7% in the derivation
Vereckei algorithm and the novel Basel algorithm in cohort and 93.3% and 90.4% in the validation cohort.
the derivation and the validation cohort, including This was similar to the Brugada and Vereckei

T A B L E 1 Patient Characteristics

Derivation Cohort Validation Cohort

All VT SVT All VT SVT

(N ¼ 124) (n ¼ 74) (n ¼ 50) P Value (N ¼ 112) (n ¼ 64) (n ¼ 48) P Value

Age, y 65 (50-74) 64 (50-72) 68 (51-77) 0.46 61 (42-69) 61 (42-70) 58 (34-69) 0.23

Male 82 (69) 51 (69) 31 (69) 0.45 64 (57) 49 (77) 30 (63) 0.14
Structural heart disease 65 (55) 51 (69) 14 (31) <0.05 56 (50) 49 (52) 7 (15) <0.05
LVEF, % 50 (34-60) 40 (27-55) 56 (50-60) <0.001 45 (30-60) 30 (25-55) 60 (51-63) <0.001
Antiarrhythmic drugs
Any 97 (78) 60 (81) 37 (74) 0.38 72 (64) 53 (83) 18 (38) <0.001
Class I 4 (3.2) 1 (1.4) 3 (6) 0.30 19 (17) 16 (25) 3 (6) 0.01
Class II 89 (72) 54 (73) 35 (70) 0.84 69 (62) 50 (78) 13 (27) <0.001
Class III 37 (30) 27 (37) 10 (20) 0.07 31 (28) 29 (45) 2 (4.2) <0.001
Class IV 4 (3.2) 1 (1.4) 3 (6) 0.30 0 0 0 1.00

Values are median (IQR) or n (%).

LVEF ¼ left ventricular ejection fraction; SVT ¼ supraventricular tachycardia; VT¼ ventricular tachycardia.
836 Moccetti et al JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022

Basel Algorithm for Diagnosis of SVT/VT JULY 2022:831–839

T A B L E 2 ECG Characteristics of Wide QRS Complex Tachycardias in the Derivation and Validation Groups

Derivation Cohort Validation Cohort

All VT SVT All VT SVT

(N ¼ 206) (n ¼ 153) (n ¼ 53) P Value (N ¼ 203) (n ¼ 151) (n ¼ 52) P Value

Cycle length, ms 380 (312-440) 376 (312-440) 400 (314-440) 0.30 405 (322-469) 402 (312-464) 413 (331-492) 0.38
Heart rate, beats/min 158 (136-192) 160 (136-192) 150 (137-191) 0.30 149 (127-186) 160 (129-191) 155 (122-185) 0.38
QRS duration, ms 160 (140-188) 172 (152-196) 140 (135-144) <0.001 165 (149-196) 182 (157-202) 155 (138-154) <0.001
RBBB 116 (58) 88 (58) 28 (58) <0.06 114 (56) 84 (56) 30 (59) 0.87
LBBB 84 (42) 64 (42) 20 (42) <0.07 89 (44) 67 (44) 22 (41) 0.11

Values are median (IQR) or n (%).

LBBB ¼ left bundle branch block; RBBB ¼ right bundle branch block; other abbreviations as in Table 1.

algorithms, in the derivation, validation, and overall [IQR: 58%-73%]; P ¼ 0.03), and no differences were
cohorts (Supplemental Table 3). The diagnostic per- found between the Basel algorithm and the other al-
formance of each step/criterion of the Brugada, Ver- gorithms (Figure 3, Central Illustration).
eckei, and novel Basel algorithm in both the Average time to diagnosis was significantly shorter
derivation and the validation cohorts are presented in using the Basel algorithm (median 38 [IQR: 2947]
Supplemental Table 4. seconds) compared with the Brugada algorithm (me-
dian 106 [IQR: 76-135] seconds; P ¼ 0.002) and Ver-
EVALUATION OF THE CLINICAL APPLICABILITY OF
eckei algorithms (median 48 [IQR: 43-59] seconds;
THE NOVEL ALGORITHM. The SN and SP of the cli-
P ¼ 0.02). No differences were found between the
nicians’ interpretation are shown in Figure 3. SP was
Basel algorithm and the other algorithms (Figure 3,
higher with the use of the Basel algorithm vs the
Central Illustration).
Vereckei algorithm (median 80% [IQR: 72%-86%] vs
median 58% [IQR: 46%-72%]; P ¼ 0.007), but no other ALGORITHM PERFORMANCE IN SPECIAL SITUATIONS. We
differences were observed between the Basel algo- tested the performance of the Basel algorithm in 3
rithm and the other algorithms. clinically challenging albeit rare scenarios: fascicular
The Basel algorithm showed a higher diagnostic ventricular tachycardia (n ¼ 3), antidromic atrioven-
accuracy (median 81% [IQR: 76.5%-83.5%]) compared tricular re-entrant tachycardia (n ¼ 1), and Mahaim-
with the Vereckei algorithm (median 72% [IQR: 65%- fiber tachycardia (n ¼ 3). Similar to the Brugada and
76%]; P ¼ 0.002) and the Chen algorithm (median 72% Vereckei algorithms, the performance of the novel
Basel algorithm was poor in these tachycardias
(Supplemental Table 5).
T A B L E 3 Underlying Causes of SVT and VT

Derivation Cohort Validation Cohort

DISCUSSION
SVT (n ¼ 53) (n ¼ 52)
Atrial flutter 22 (42) 21 (40)
By analyzing data from 2 large cohorts of patients
AVNRT 17 (34) 18 (35)
AVRT 4 (8) 3 (6)
with EP-confirmed diagnosis, we developed and
Atrial tachycardia 8 (15) 8 (15) validated a novel, simple, reproducible, sensitive,
Mahaim 2 (4) 1 (2) and specific algorithm based on clinical and ECG pa-
Antidromic AVRT 0 (0) 1 (2) rameters to discriminate VT from SVT in patients with
VT (n ¼ 153) (n ¼ 151) regular monomorphic WCT. Compared with the cur-
Ischemic 93 (63) 65 (43)
rent standard of care,10,13 the proposed Basel algo-
DCM 32 (21) 49 (32)
rithm provides similarly high diagnostic accuracy and
Other 28 (18) 37 (25)
performance characteristics while at the same time
ARVC 5 (3) 4 (3)
RVOT 9 (6) 3 (2) offering a simplified approach based on clear-cut
Fascicular 2 (1) 1 (0.7) easy-to-use criteria and by avoiding measurements
HCM 1 (0.7) 1 (0.7) of complex and morphology-based parameters that
are difficult to assess in the emergency setting (eg, V i /
Values are n (%).
V t).23 Since a common shortcoming for the Brugada
ARVC ¼ arrhythmogenic right ventricular cardiomyopathy; AVNRT ¼ atrioven-
tricular non-reentrant tachycardia; AVRT ¼ atrioventricular re-entrant tachycardia; and Vereckei algorithms is their lower accuracy in the
DCM ¼ dilated cardiomyopathy; HCM ¼ hypertrophic cardiomyopathy;
RVOT ¼ right ventricular outflow tract; other abbreviations as in Table 1.
“real-world” setting compared with the performance
described in original publications (Supplemental
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022 Moccetti et al 837
JULY 2022:831–839 Basel Algorithm for Diagnosis of SVT/VT

Table 1), 10-12,18,19,23 we sought to further validate the

T A B L E 4 Diagnostic Performance Characteristics of the Brugada, 10
Basel algorithm in a clinical setting. When applied by Vereckei, 12 and Basel Algorithms
clinicians with different training/background, in a
Diagnostic
head-to-head comparison against the Brugada, Sensitivity Specificity PPV NPV Accuracy
Vereckei, and 3 other previously published algo- Derivation cohort (n ¼ 206)
rithms,14-16 the novel Basel algorithm showed a Brugada 92.8 90.6 96.6 81.4 92.2
diagnostic performance similar to that of the Brugada Vereckei 96.7 86.8 95.5 90.2 94.2
algorithm and higher than that of the Vereckei algo- Basel 91.5 88.7 95.9 78.3 90.8

rithm, while requiring significantly shorter time for Criterion I (clinical) 73.9 69.8 87.6 48.1 72.8
Criterion II (lead II) 86.9 94.3 97.8 71.4 88.8
diagnosis. This difference in time was most pro-
Criterion III (lead aVR) 79.1 92.5 96.8 60.5 82.5
nounced when applied by cardiology fellows and in-
Validation cohort (n ¼ 203)
ternal medicine residents. Such performance Brugada 93.3 88.5 95.9 82.1 92.1
characteristics indicate that the low complexity and Vereckei 91.3 84.6 94.5 77.2 89.6
high accuracy of the Basel algorithm make it highly Basel 93.3 90.4 96.6 82.5 92.6
applicable in clinical practice and particularly useful Criterion I (clinical) 84.7 80.8 92.7 64.4 83.7
for physicians in training to ensure timely diagnosis Criterion II (lead II) 89.3 84.6 94.4 73.3 88.1
Criterion III (lead aVR) 86.0 96.1 98.5 70.0 88.6
of WCT.
Similar to the lead aVR (Vereckei) algorithm13 and
Values are %.
the R-wave peak time (RWPT) algorithm,14 the Basel NPV ¼ negative predictive value; PPV ¼ positive predictive value.
algorithm is based on the analysis of the electrical
vector in the frontal plane only. However, the afore-
mentioned algorithms operate based on only 1 lead
(leads aVR or II), with the diagnosis of WCT primarily CLINICAL IMPLICATIONS. The Basel WCT algorithm

determined by R-wave parameters. Alongside clinical can be used for a quick and accurate differential
parameters highly sensitive for VT, the Basel algo- diagnosis of WCT by EP and non-EP practitioners. Of
rithm offers an integrated yet simplified and accurate note, the focus on leads aVR and II in the Basel al-
version of 2 ECG criteria from the 2 aforementioned gorithm make it potentially suitable for the use in
algorithms: RWPT >40 ms in lead aVR, which is the emergency settings and for incorporation in Holter
second step of the Vereckei algorithm, 13
and a posi- and telemetry analysis software that include infor-
tive or negative RWPT $50 ms in lead II, the criterion mation on only limb leads.
14
behind the RWPT algorithm. The theory behind STUDY LIMITATIONS. The findings of this study
both criteria is that SVT with bundle branch block should be viewed in the light of a few potential limi-
results in rapid initial activation of the ventricular tations. 1) The derivation and validation cohorts
myocardium owing to impulse conductance through comprised patients from tertiary EP centers, so referral
the His-Purkinje system (steeper QRS in both aVR and bias cannot be excluded. 2) All patients included in this
II leads), with a delayed muscle-to-muscle spread of study had EP-confirmed diagnoses. While this meth-
activation, resulting in widening of the terminal QRS. odology allows having a criterion standard for diag-
In contrast, during VT, initial ventricular activation is nosing the mechanism of WCT, a selection bias might
through muscle-to-muscle spread of the impulse, be present because not all patients with WCT undergo
which is slower than terminal activation mediated an EP study. 3) Both cohorts contained a relatively low
through the His-Purkinje system after the impulse number of cases with preexcitation-related tachy-
reaches the conduction system. This results in cardia. Although pre-excited SVTs are a rare cause of
steeper terminal QRS. WCT,13 the Basel algorithm needs to be further studied
It should be recognized that identifying the initia- in those cases. 4) For some of the patients, more than 1
tion, peak, and termination of the QRS complex might VT ECG was included. These were, however, ECGs from
at times be challenging for any given single lead, different VTs with different ECG morphologies. 5)
particularly for aVR13; thus, integration of 2 limb leads Neither our approach nor the available algorithms are
in the Basel algorithm will likely help in determining perfect. Given the rapid implementation of techno-
the time to first peak in either of the leads, when logic advances in clinical care, development of an
measurements in single leads are difficult. Further automated application that uses a complex WCT dif-
prospective studies comparing the Basel algorithm ferential algorithm, with incorporation of the Basel
with other proposed criteria are needed to determine and or other WCT algorithm, might enable a more ac-
the validity of our criteria in distinguishing between curate and quicker differential diagnosis of WCT in the
VT and SVT with wide QRS complexes. near future.
838 Moccetti et al JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022

Basel Algorithm for Diagnosis of SVT/VT JULY 2022:831–839

F I G U R E 3 Clinical Validation of the Basel Algorithm

Comparison of sensitivity, specificity, diagnostic accuracy, and average time for electrocardiographic analysis of the Basel algorithm with the
Vereckei,13 Brugada,10 Jastrzebski,15 Pava,14 and Chen16 algorithms based on a subset of the derivation cohort and stratified by the observers’
specialties. EP ¼ electrophysiology; IM ¼ internal medicine.

CONCLUSIONS reached in a shorter time, particularly when applied

by physicians in training.
We constructed an easy-to-use and accurate algo- FUNDING SUPPORT AND AUTHOR DISCLOSURES
rithm to distinguish VT from SVT with aberrant con-
duction with the use of a 6-lead limb ECG. The Basel Dr Schaer has received personal fees from Medtronic. Dr Kühne has
received grants from the Swiss National Science Foundation, the Swiss
algorithm consists of 3 criteria and had a performance
Heart Foundation, Daiichi-Sankyo, Bayer, Pfizer BMS, and Boston
similar to that of the established Brugada and Ver- Scientific; and has received personal fees from Bayer, Boehringer
eckei algorithms in both derivation and validation Ingelheim, Pfizer BMS, Daiichi-Sankyo, Medtronic, Biotronik, Boston
cohorts. When applied by physicians with different Scientific, and Johnson & Johnson, all outside the submitted work. Dr
Henrikson has received fellowship support from Abbott, Boston Sci-
training levels and backgrounds, the novel algorithm
entific, and Medtronic; and has served as chair of the clinical endpoints
showed a high accuracy similar to the Brugada and committee for Biotronik. Dr Sticherling has received grants from
Vereckei algorithms while allowing a diagnosis to be Biosense-Webster; and has received lecture fees from Abbott,
JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 8, NO. 7, 2022 Moccetti et al 839
JULY 2022:831–839 Basel Algorithm for Diagnosis of SVT/VT

Medtronic, Biosense-Webster, Boston Scientific, Microport, and Bio-

PERSPECTIVES
tronik. Dr Reichlin has received speaker/consulting honoraria or travel
support from Abbott/SJM, AstraZeneca, Brahms, Bayer, Biosense-
Webster, Biotronik, Boston-Scientific, Daiichi-Sankyo, Medtronic,
COMPETENCY IN MEDICAL KNOWLEDGE: The novel
Pfizer BMS, and Roche, all for work outside the submitted study; and
has received support for his institution’s fellowship program from
algorithm can be used for rapid and accurate differential diag-
Abbott/SJM, Biosense-Webster, Biotronik, Boston-Scientific, and nosis of wide QRS complex tachycardias by physicians with
Medtronic for work outside the submitted study. All other authors have different backgrounds and training levels.
reported that they have no relationships relevant to the contents of this
paper to disclose.

TRANSLATIONAL OUTLOOK :The development of auto-

ADDRESS FOR CORRESPONDENCE: Prof Tobias
mated applications using combinations of complex algorithms
Reichlin, Department of Cardiology, Inselspital, Bern
might further improve the differential diagnosis of wide QRS
University Hospital, University of Bern, Freiburg-
complex tachycardias.
strasse 10, CH-3010 Bern, Switzerland. E-mail: tobias.
[email protected].

REFERENCES

1. Akhtar M, Shenasa M, Jazayeri M, Caceres J, 10. Brugada P, Brugada J, Mont L, Smeets J, medicine residents to distinguish between ven-
Tchou PJ. Wide QRS complex tachycardia. Reap- Andries EW. A new approach to the differential tricular tachycardia and supraventricular tachy-
praisal of a common clinical problem. Ann Intern diagnosis of a regular tachycardia with a wide QRS cardia with aberrancy. J Cardiol. 2012;59:307–
Med. 1988;109:905–912. complex. Circulation. 1991;83:1649–1659. 312.

2. Steinman RT, Herrera C, Schuger CD, 11. Lau EW, Pathamanathan RK, Ng GA, Cooper J, 20. Isenhour JL, Craig S, Gibbs M, Littmann L,
Lehmann MH. Wide QRS tachycardia in the Skehan JD, Griffith MJ. The Bayesian approach Rose G, Risch R. Wide-complex tachycardia:
conscious adult. Ventricular tachycardia is the improves the electrocardiographic diagnosis of continued evaluation of diagnostic criteria. Acad
most frequent cause. JAMA. 1989;261:1013–1016. broad complex tachycardia. Pacing Clin Electro- Emerg Med. 2000;7:769–773.
physiol. 2000;23:1519–1526.
3. Page RL, Joglar JA, Caldwell MA, et al. 2015 21. Szelényi ZDG, Katona G, Fritúz G, et al.
ACC/AHA/HRS guideline for the management of 12. Vereckei A, Duray G, Szenasi G, Altemose GT, Comparison of the “real-life” diagnostic value of
adult patients with supraventricular tachycardia: Miller JM. Application of a new algorithm in the two recently published electrocardiogram
a report of the American College of Cardiology/ differential diagnosis of wide QRS complex methods for the differential diagnosis of wide
American Heart Association Task Force on Clin- tachycardia. Eur Heart J. 2007;28:589–600. QRS complex tachycardias. Acad Emerg Med.
ical Practice Guidelines and the Heart Rhythm 2013;20:1121–1130.
13. Vereckei A, Duray G, Szenasi G, Altemose GT,
Society. J Am Coll Cardiol. 2016;67(16):1575– Miller JM. New algorithm using only lead aVR for 22. Herbert MEVS, Morgan MT, Cameron P,
1623. differential diagnosis of wide QRS complex Dziukas L. Failure to agree on the electrocardio-
tachycardia. Heart Rhythm. 2008;5:89–98. graphic diagnosis of ventricular tachycardia. Ann
4. Al-Khatib SM, Stevenson WG, Ackerman MJ,
Emerg Med. 1996;27:35–38.
et al. 2017 AHA/ACC/HRS guideline for manage- 14. Pava LF, Perafan P, Badiel M, et al. R-Wave
ment of patients with ventricular arrhythmias and peak time at DII: a new criterion for differentiating 23. Kaiser E, Darrieux FC, Barbosa SA, et al. Dif-
the prevention of sudden cardiac death: a report between wide complex QRS tachycardias. Heart ferential diagnosis of wide QRS tachycardias:
of the American College of Cardiology/American Rhythm. 2010;7:922–926. comparison of two electrocardiographic algo-
Heart Association Task Force on Clinical Practice rithms. Europace. 2015;17:1422–1427.
15. Jastrzebski M, Sasaki K, Kukla P, Fijorek K,
Guidelines and the Heart Rhythm Society. J Am
Stec S, Czarnecka D. The ventricular tachycardia
Coll Cardiol. 2018;72(14):e91–e220.
score: a novel approach to electrocardiographic
5. Alzand BS, Crijns HJ. Diagnostic criteria of broad diagnosis of ventricular tachycardia. Europace.
KEY WORDS ECG, cardiac arrhythmia, wide
QRS complex tachycardia: decades of evolution. 2016;18:578–584.
QRS complex tachycardia, ventricular
Europace. 2011;13:465–472. 16. Chen Q, Xu J, Gianni C, et al. Simple electro- tachycardia, supraventricular tachycardia,
6. Stewart RB, Bardy GH, Greene HL. Wide com- cardiographic criteria for rapid identification of sudden cardiac death, algorithm
plex tachycardia: misdiagnosis and outcome after wide QRS complex tachycardia: the new limb lead
emergent therapy. Ann Intern Med. 1986;104: algorithm. Heart Rhythm. 2020;17:431–438.
766–771.
17. Pachon M, Arias MA, Salvador-Montanes O,
7. Pelleg A, Pennock RS, Kutalek SP. Proar- et al. A scoring algorithm for the accurate differ- A PP END IX For supplemental tables and a
rhythmic effects of adenosine: one decade of ential diagnosis of regular wide QRS complex figure, please see the online version of this
clinical data. Am J Ther. 2002;9:141–147. tachycardia. Pacing Clin Electrophysiol. 2019;42: paper.
625–633.
8. Buxton AE, Marchlinski FE, Doherty JU,
Flores B, Josephson ME. Hazards of intravenous 18. Jastrzebski M, Kukla P, Czarnecka D,
verapamil for sustained ventricular tachycardia. Kawecka-Jaszcz K. Comparison of five electro-
Am J Cardiol. 1987;59:1107–1110. cardiographic methods for differentiation of wide Go to http://www.jac-
QRS-complex tachycardias. Europace. 2012;14: c.org/cme-moc to take
9. Garner JB, Miller JM. Wide complex tachy- the CME/MOC/ECME quiz
1165–1171.
cardia—ventricular tachycardia or not ventricular for this article.
tachycardia, that remains the question. Arrhythm 19. Baxi RP, Hart KW, Vereckei A, et al. Vereckei
Electrophysiol Rev. 2013;2:23–29. criteria as a diagnostic tool amongst emergency