Blind Penetration or Pace‐Probing of the Septal Substrate for Left Bundle Branch Pacing?

With His bundle pacing, the target area for lead implantation can easily be determined by localizing the His bundle potential, or better still, by pacemapping to identify the area where a narrow physiological QRS can be obtained at a moderate pacing output. The success of His bundle lead implantation is obvious immediately upon deployment, depending mainly on pinpointing the identified target with the active helix of the pacing lead. The septal substrate is not important for His bundle pacing. The story is completely different with left bundle branch (LBB) area pacing. The current LBB area pacing technique involves a more or less blind attempt to penetrate the septum with the pacing lead, with the target area for lead deployment only approximated by anatomical markers. The operator undertaking a blind penetration attempt does not know if the lead will pass through the septum at the chosen location, be entangled by the initial endocardial barrier [1], or be stopped by deep intraseptal fibrosis. Moreover, the operator does not even know if there is a LBB on the other side of the septum at the chosen location, or if it is just an empty space between the major fascicles of the LBB. Furthermore, the operator does not know the functional condition of the left conduction system. In patients with heart failure, we often capture the LBB, yet the paced QRS remains far from physiological—a condition that we have termed functional deep septal pacing (DSP) [2]. Functional DSP results from latency and/or slow conduction in the LBB and is characterized by a broad QRS complex that lacks a terminal R wave in V1 and has a long R wave peak time in lead V6 (V6-RWPT)—Figure 1. To achieve electrical and/or mechanical resynchronization in patients with functional DSP, it appears essential to add a coronary venous pacing lead and utilize either left bundle branch-optimized (LOT)-CRT or conventional CRT [3, 4]. To summarize, given the known variability of LBB anatomy, the variable functional condition of the left conduction system and the variable challenges imposed by the septal substrate, it seems fitting to paraphrase: “The LBB in a CRT candidate is like a box of chocolates. You never know what you′re gonna get when you target it.” The initial attempts to make the left bundle branch (LBB) target more predictable focused on improving the characterization of septal substrates by identifying septal fibrosis using late gadolinium enhancement cardiac magnetic resonance imaging [5, 6]. Indeed, the presence of fibrosis correlates to some degree with implantation failure, and some clinicians even recommend directing CRT candidates with high septal scar burden empirically towards conventional biventricular pacing rather than LBB area pacing. However, reasoning based on septal scar seems quite limited to me. This is because the presence of a septal scar does not indicate the functional condition of the LBB, which is more important. While septal scar may not be a problem in the near future when RF-current facilitated lead implantation enters clinical practice, a good functional condition of the LBB will remain the sine qua non of LBB-based CRT. In the current issue of JCE, Cai et al. [7] proposed a different method for septal substrate assessment, a method that theoretically may provide information on both the condition of the septal substrate and the functional state of the left conduction system. The authors retrospectively analyzed the value of the V6 R-wave peak time (RWPT), obtained during initial right ventricular septal pacing (RVSP), as a predictor of successful LBB area pacing in 99 CRT candidates with left bundle branch block. They found that RVSP V6-RWPT < 145 ms was a better predictor of success than baseline left ventricular end-diastolic dimension and QRS duration—risk markers for LBB area pacing failure previously established by MELOS [8]. This interesting finding begs the question: what determines V6-RWPT during RVSP? The depolarization wavefront initiated by RVSP first has to cross the septal working myocardium via slow cell-to-cell conduction. This takes about 40–50 ms in healthy myocardium and often 70-80 ms or more in heart failure patients. Then the depolarization wavefront enters retrogradely the left conduction system via nearest myocardium-to-Purkinje network junction and depolarizes rest of the left ventricle. These two components seem critical—the transseptal conduction time (reflecting the septal substrate) and functional condition of the LBB (impacting the degree of latency and velocity of conduction) - Figure 2. Indeed, it was found by Cai et al. that septal scar burden correlated with RVSP V6-RWPT. What we lack in this study, to better understand the physiological background of RVSP V6-RWPT, is some correlation with functional status of the left conduction system (e.g. with the latency interval observed during selective LBB capture). However, studies that analyzed V6-RWPT values for capture type differentiation showed that, despite LBB capture, the paced V6-RWPT values are much higher in patients with baseline diseased left conduction system versus patients with baseline narrow QRS or isolated right bundle branch block [9]. Therefore, it is safe to assume that functional condition of the left conduction system contributes also to RVSP V6-RWPT. While using V6 RWPT during RVSP as measure for pace probing the septum is an interesting idea, the study by Cai et al. is not without limitations. Firstly, the value of the RVSP V6-RWPT as a predictor of success depends critically on how LBB pacing success in a CRT candidate is defined. The definition used by the authors may not stand the test of time. For example, their arbitrary definition of acute success—obtaining LBB area capture with 20% reduction in QRS width may not equate to successful CRT. The value of the study would be much higher if correlation of RVSP V6-RWPT with long-term clinical and echocardiographic response is provided. Secondly, a specificity of around 90% means that in 10% of patients despite RVSP V6-RWPT > 145 ms successful LBB-CRT can still be obtained. In my view, this limits the practical usefulness as it would be difficult to refuse an attempt at LBBAP when you know it may still be successful in a significant percentage of patients. Thirdly, the reported RVSP values, 161 ms and 132 ms, in successful and failed LBB area pacing, respectively, differ from that observed in other cohorts. For instance, in our study that analyzed paced V6-RWPT at four functional depths in the septum, an average RVSP V6-RWPT was much shorter (117 ms) [10]. In summary, Cai and his colleagues should be commended for advancing the field of LBB pacing and presenting hypothesis-generating study. However, further research is required, particularly in other populations, before the cut-off point of 145 ms can be recommended for clinical use. The author declares no conflicts of interest.