Differentiating the QRS Morphology of Posterior Fascicular Ventricular Tachycardia From Right Bundle Branch Block and Left Anterior Hemiblock Aberrancy

Authors, Journal, Affiliations, Type, DOI

• Authors: Yoav Michowitz, Oholi Tovia-Brodie, Ishai Heusler, Avi Sabbag, Michael Rahkovich, Hezzy Shmueli, Aharon Glick, Bernard Belhassen
• Journal: Circulation: Arrhythmia and Electrophysiology, 2017;10:e005074
• Affiliations: Department of Cardiology, Tel Aviv Sourasky Medical Center and Sheba Medical Center, Israel; Sackler Faculty of Medicine, Tel Aviv University
• Type: Original article (retrospective case-control with Medline case collection)
• DOI: https://doi.org/10.1161/CIRCEP.117.005074

Overview

The first systematic ECG characterisation of left posterior fascicular VT (LPF-VT) against its principal mimic — SVT with RBBB plus left anterior hemiblock (LAHB). Using 183 EP/ablation-confirmed LPF-VT ECGs versus 61 RBBB+LAHB sinus rhythm controls, multivariate logistic regression identified four independent ECG discriminators: atypical V1 morphology (OR 5.1), positive QRS in aVR (OR 19.2), V6 R/S ≤1 (OR 6.7), and QRS ≤140 ms (OR 7.7), yielding Sn 82.1%/Sp 78.3%. Critically, the paper confirms that all standard WCT algorithms (Brugada RS >100 ms, Vereckei Vi/Vt, Wellens >140 ms) fail for LPF-VT because Purkinje-mediated initial activation is fast — RS interval is <80 ms and Vi/Vt is typically >1, the reverse of structural VT. This paper fills the documented fascicular VT blind spot in all existing WCT algorithms.

Keywords

Bundle branch block, electrocardiography, logistic models, odds ratio, tachycardia ventricular, left posterior fascicular VT, RBBB, left anterior hemiblock, verapamil-sensitive VT

Key Takeaways

Background

• LPF-VT = left posterior fascicular VT = most common form of fascicular VT (~80% of fascicular VTs); mechanism is macroreentry in the Purkinje network involving the posterior fascicle
• Also called "Belhassen VT," "idiopathic left VT," or "verapamil-sensitive VT"
• Frequently misdiagnosed as SVT with RBBB+LAHB aberrancy because: (1) occurs in young patients without structural heart disease; (2) almost always haemodynamically well-tolerated; (3) terminated by IV calcium channel blockers (verapamil) — mimicking adenosine-sensitive SVT response
• All standard WCT algorithms fail because LPF-VT activation exits via the left posterior Purkinje fascicle → initial rapid activation (not slow muscle-to-muscle):
  • RS interval <80 ms (Andrade et al. 1996; confirmed in this study: median 62 ms) — below Brugada >100 ms threshold
  • Vi/Vt typically >1 (fast initial, slow terminal) → Vereckei algorithm classifies as SVT
  • QRS ≤140 ms (mean 127.5 ms) — Wellens' structural VT criterion of >140 ms is INVERTED: RBBB+LAHB controls had wider QRS (mean 144 ms)
  • Q wave in aVR <40 ms — Vereckei initial-R-in-aVR criterion requires initial positive deflection; LPF-VT typically has qR complex (narrow q <40 ms before R), so the Vereckei aVR criterion is negative

Methods

• LPF-VT group (n=183 ECGs):
  • 39 patients who underwent RF ablation at Tel Aviv/Sheba Medical Centers (1992–2013)
  • 144 ECGs retrieved from Medline search (87 publications, 1981–2013); included only if EP study + ablation confirmed LPF-VT, no organic heart disease, ECG quality adequate; case reports included, review articles excluded
  • Publication bias tested: no significant differences between ablation group and Medline group
• Control group (n=61 ECGs): consecutive patients with RBBB+LAHB on ECG, no significant structural heart disease (echo-confirmed), normal LV function; patients in sinus rhythm
  • LAHB defined: QRS axis ≤−30° + small r followed by deep S in inferior leads
• Atrial pacing control group (n=16): patients with RBBB+LAHB who underwent EP study and had atrial pacing at ≥100 bpm — confirms rate-independent ECG features
• Two blinded observers with consensus; interobserver κ >0.8 for all parameters

ECG Findings — Univariate Comparisons (Table)

• Precordial RS time: NOT significantly different between LPF-VT and RBBB+LAHB (P=0.4) — both <80 ms; confirms this criterion has no discriminatory value for LPF-VT

The Four-Variable Prediction Model

Multivariate logistic regression identified four independent predictors (probability equation: P = 1 / [1 + e^−Z], where Z = −5.219 + contributions from each variable):

1. Atypical V1 morphology (OR 5.1; 95% CI 1.7–15.6; P=0.004; +1.636 to Z)

• Definition: R′ is NOT taller than R (absence of typical rsR′); includes qR, RsR′ with R≈R′ (notched), or monophasic R
• Present in 45.8% of LPF-VT vs 8.2% of controls
• Note: 54.2% of LPF-VT still had TYPICAL V1 morphology (rsR′ with R′>R) — atypicality supports VT but typical V1 does NOT exclude it
• Additional V1 features supporting LPF-VT (not in model): q wave before rsR′ (35.1% vs 7.1%); notched V1 (24.4% vs 1.6%)
• S wave below isoelectric line supports RBBB+LAHB aberrancy (72.1% vs 21.1%)

2. Positive QRS in aVR (OR 19.2; 95% CI 4.3–86.5; P<0.001; +2.958 to Z)

• Definition: R>S or R>Q (overall QRS polarity positive in aVR); this is broader than Vereckei's "initial R" criterion
• Present in 94.2% LPF-VT vs 50.8% RBBB+LAHB
• Note: RBBB+LAHB often produces northwest axis → more frequent positive aVR than RBBB alone — this reduces the discriminating power compared with structural VT vs SVT
• Most LPF-VT show qR complex in aVR (narrow q <40 ms → Vereckei initial-R criterion NOT triggered despite positive overall QRS)
• Sex difference: men 98.8% positive aVR vs women 88% (P=0.04) due to more negative axis in men

3. V6 R/S ratio ≤1 (OR 6.7; 95% CI 1.6–28.5; P=0.01; +1.901 to Z)

• Present in 88.3% LPF-VT vs 59% RBBB+LAHB (considerable overlap)
• Pathophysiology: in LPF-VT, activation spreads away from V6 through left ventricular myocardium → deep S; in RBBB aberrancy, delayed RV activation creates small S; BUT addition of LAHB deepens S in V6 even in aberrancy → overlap increases
• Single discriminator: R/S ratio ≤0.15 in V6 seen ONLY in LPF-VT (21.9%)

4. QRS ≤140 ms (OR 7.7; 95% CI 2.9–20.3; P<0.001; +2.040 to Z)

• Mean QRS 127.5±18.6 ms (LPF-VT) vs 144±15.4 ms (RBBB+LAHB)
• This INVERTS the Wellens structural VT criterion (QRS >140 ms): RBBB+LAHB is WIDER than LPF-VT
• Single discriminators (seen only in LPF-VT): QRS <120 ms (31% of LPF-VT); QRS axis <−100° (16.7%)

Model Performance and Application Rules

Overall performance (threshold probability >0.59):

• Sensitivity 82.1%, Specificity 78.3%, PPV 85.7%, NPV 73.4%, Accuracy 80.6%
• AUC 0.889 (95% CI 0.836–0.943); good calibration (Hosmer-Lemeshow P=0.812)
• 5- and 10-fold cross-validations confirmed same sensitivity and specificity

Practical counting rule (Figure 2B):

• 0–1 positive criteria: RBBB+LAHB aberrancy (definite/probable) — LPF-VT virtually excluded
• 2 positive criteria: probable LPF-VT
• 3–4 positive criteria: LPF-VT (high probability)

Proposed clinical algorithm (Figure 5):

1. AV dissociation or fusion/capture beats present → VT confirmed
2. Structural heart disease present → apply standard WCT algorithms (Brugada, Vereckei, etc.)
3. No structural disease, RBBB+LAHB-like morphology → apply 4-criterion model above

LPF-VT vs Posterior Papillary Muscle VT (PPM-VT) — Differential

• PPM-VT: NOT verapamil-sensitive; focal mechanism (not reentry); QRS >160 ms (vs 127 ms LPF-VT); variable spontaneous QRS morphologies; PPM-VT from Komatsu 2017 — R/S <1 in lead I
• Overlap: QRS 127–142 ms overlap zone; R/S <1 in lead I present in 33.5% of LPF-VT
• Most reliable single distinguisher: QRS >160 ms → PPM-VT (present in only 4.4% of LPF-VT)
• Typical rsR′ in V1 and discrete Q in leads I/aVL support LPF-VT but do not exclude PPM-VT

Why Standard WCT Algorithm Criteria Fail for LPF-VT

Limitations of the Document

• Retrospective design: no prospective validation of prediction model
• Mixed case source: 144/183 LPF-VT ECGs from published literature — publication bias possible (authors compared ablation vs Medline groups and found no significant differences, but cannot exclude)
• Control group not in tachycardia: RBBB+LAHB controls were mostly in sinus rhythm (not pacing-induced aberrancy); atrial pacing control group was small (n=16) and at lower rates than LPF-VT cases
• Control group age mismatch: controls were older (median ~76 years) vs LPF-VT (median 30 years) — age effect on ECG morphology not fully controlled
• Applies only to LPF-VT (~80% of fascicular VTs): left anterior fascicular VT (LAF-VT; RBBB+right axis) and upper septal fascicular VT (narrow QRS) not analysed
• Sex: reduced discriminating power in women (positive aVr 88% vs 99% men)
• Overlap zone at 2 positive criteria: probability is not deterministic; near-threshold QRS width values especially problematic
• No independent validation cohort: cross-validation is internal; external validation needed

Key Concepts Mentioned

• concepts/Fascicular-Ventricular-Tachycardia — primary subject; this paper provides the first systematic ECG characterisation of LPF-VT
• concepts/Wide-Complex-Tachycardia — all standard WCT algorithms fail for LPF-VT; this paper documents specific reasons and provides an alternative approach

Key Entities Mentioned

• None prominent beyond the arrhythmia entity itself

Wiki Pages Updated

• wiki/sources/lpfvt-svt-circep-2017.md — created (this file)
• wiki/concepts/Fascicular-Ventricular-Tachycardia.md — created
• wiki/concepts/Wide-Complex-Tachycardia.md — fascicular VT section enriched, source link added, source_count 6→7
• wiki/sourceindex.md — added entry
• wiki/wikiindex.md — added Fascicular-Ventricular-Tachycardia entry; updated Wide-Complex-Tachycardia entry