Transformation of a tachycardia following a His‐refractory premature ventricular complex: What is the mechanism?

A 69-year-old female underwent an electrophysiological study for recurrent supra-ventricular tachycardia (SVT). Programmed electrical stimulation from the high right atrium reproducibly induced a narrow complex tachycardia. During one of the induction attempts, a wide complex tachycardia (WCT) was seen (Figures 1 and 2). A ventricular extra stimulus introduced during His bundle refractoriness changed the morphology of the tachycardia (Figures 3 and 4). What is the mechanism of the tachycardia? How can you explain the morphology change? The atrial extra stimulus initiates a WCT (QRS 150 ms) with atrial-His (AH) prolongation. The QRS has a left bundle branch block (LBBB) pattern. The atrial activation pattern is concentric, with the earliest atrial signal at the His electrode. The earliest ventricular signal is at the right ventricular apex electrode. The tachycardia's cycle length (CL) is regular (305–315 ms). It is a "V-on-A" tachycardia. Given the initiation with sudden AH prolongation and septal ventriculoatrial (VA) time of <70 ms, the diagnosis of typical slow-fast atrioventricular nodal reentrant tachycardia (AVNRT) can already be made at this point.1 The wide complexes demonstrate a short HV interval, suggesting ventricular pre-excitation. Closer analysis shows subtle changes in the QRS morphology with corresponding changes in HV interval (Figure 2). The PVC is delivered during His-refractoriness, as suggested by its timing relative to the next expected His and the fused QRS complex. Despite narrowing the QRS (80 ms), the tachycardia is unperturbed and maintains a similar CL, atrial activation sequence, and AH relationships. Notably, the immediate atrial signal occurs at the expected timing. The short HV interval suggests anterograde conduction through an accessory pathway (AP). The wide QRS of LBBB morphology results from a varying degree of fusion between anterograde conduction through the atrioventricular (AV) node and AP, as suggested by the HV interval variation. The "long" AV interval is not classical for a typical AP. All of the above mentioned are in fact features of an atrio-fascicular AP (AFP).2 Loss of pre-excitation following a His-refractory PVC, proves concealed retrograde block in the AP. Whereas AFPs typically lack retrograde conduction to the atrium, concealed retrograde block in the AP is still possible. Close attention to the QRS following the first atrial extrastimulus starting the tachycardia (Figure 1), shows a likely fusion between AP conduction and AV nodal conduction (relatively short AH although the His signal is subtle). The stim-QRS is short and would be unexpected, although not impossible, with an AFP. Therefore, this is, in all likelihood, a bidirectional conducting nodofascicular/nodoventricular (NFV) AP acting as a bystander in the setting of typical AVNRT. NFVs are rare APs that share common characteristics with AFPs but have different anatomical locations.3 They were initially described by Mahaim and Winston as APs with decremental conduction properties.4 As opposed to "typical" AFPs that run along the lateral tricuspid annulus area, NFV connect the AV node to the right ventricle or right bundle. They commonly participate as the retrograde limb in narrow complex tachycardias (i.e. orthodromic atrioventricular reentrant tachycardia AVRT). Still, they may also cause WCT by serving as the anterograde limb of the reentry circuit (i.e. antidromic AVRT). They may as well be present without overt tachycardia or as a bystander in SVT. Empirical ablation at the typical slow pathway location is often sufficient to eliminate these APs. Radiofrequency ablation anterior to coronary sinus ostium, at the base of the triangle of Koch, rendered AVNRT or any other SVT non-inducible and eliminated the AP in this case. The repeat EP study was normal without any evidence of pre-excitation or persistence of an AP, furthermore suggesting a NFV, which must have been concomitantly eliminated. The authors declare no conflict of interest. Data sharing is not applicable to this article as no new data were created or analyzed in this study.