In situ anatomy map provides a new scenario for conduction system pacing

Complex current diagnostic and treatment techniques necessitate new anatomical maps of the cardiac conduction system (CCS) that are accurate in terms of arbitrary field of view and relative sizing. Recently, conduction system pacings (CSPs), such as His bundle pacing (HBP) and left bundle branch pacing (LBBP), has emerged as an option to replace conventional right ventricular apical and biventricular pacing. However, it is still impossible to visualize CCS morphology using available clinical imaging modalities. With reference to anatomical landmarks in fluoroscopic imaging, implantation is performed based on intracardiac electrogram potentials. Some gold standard manuals for successful pacing have been published to assist with highly technical procedures and to prevent loss of capture.1,2 However, the CCS has individual variation and complex localization embedded within the heart.3 Additionally, it is technically difficult to implant a wire in the His bundle (HB), measuring a maximum of 2 mm, and the left bundle branch (LBB), measuring <1 mm. Some excellent anatomical studies of the CCS using recent technical advances have revealed detailed three-dimensional (3D) CCS architecture.4,5 However, the 3D data obtained from orientation-lost ex vivo hearts are not useful as a 3D anatomic navigator for CSP. The century-old conventionally used anatomical model of the CCS is associated with many misunderstandings and misdescriptions6,7 and is not adaptable to CSP for two-dimensional (2D), simplified models. In the common diagram used in most clinical electrophysiological papers, the HB runs alone and then bifurcates into the right bundle branch (RBB) and LBB. In other words, only HBP is unambiguously implanted in the HB, and the catheter lead of the LBBP always crosses the RBB6,7 (Figure 1A). It is well-known that the RBB is a tiny muscular bundle and the LBB is a thin and broad structure (Figure 1B). Some cardiologists believe that as both bundle branches bifurcate but are depicted in the same plane in the 2D diagram, the RBB is occasionally injured during LBBP. Su et al. reported that among 638 LBBPs, 618 (97.8%) were successful, while permanent RBB injury occurred in 55 (8.9%),8 clearly indicating that an accurate 3D anatomical navigation map is needed. (A) The alternatively and conventionally used anatomical model of the CCS for over a century represents that HBP is unambiguously implanted only in the HB, and the catheter lead of the LBBP always crosses the RBB. However, the photograph (B) and the in situ 3D model (C) of in situ CCS shows that (1) HBP usually implants not only HB but also initial LBB simultaneously at the location of the thin interventricular muscle at the anteroinferior margin of the membranous septum and (2) no RBB injury during LBBP that implants perpendicular to the interventricular septum muscles at 10–15 mm inferior from the HB capture, as indicated by detailed pacing sites based on anatomic (D) and histologic (E) evidence. Our novel 3D model of in situ CCS9 emphasized that the LBB and RBB do not diverge into two branches, but rather that the LBB diverges from the branching HB as a wide and thin sheet, and then forms the RBB after branching off the LBB completely. While this finding itself is not novel, the large field of view, including surrounding structures, along with accurate in situ 3D CCS localization provided the model with minimal deformation and shrinkage due to technical preparation. Based on this technique, we present the first simulation of CSP using the right anterior oblique (RAO) 30° projection to evaluate the pacing implantation site and the left anterior oblique (LAO) 30–45° projection for perpendicular implantation in the interventricular septal muscle for preventing loss of capture (Figure 1C). In situ 3D models are less prone to RBB injury during LBBP. Anatomically, an LBBP that implants perpendicular to the interventricular septal muscle at 10–15 mm inferiorly from the HB capture or the tricuspid valve annulus10 are not likely to intersect or damage the RBB. The proximal LBB and RBB start in different positions and distribute in different directions; therefore, if the LBBP is implanted in this proximal position, the LBB and RBB are separated in any plane and cannot be damaged, unless the implantation occurs more distally in the LBB or at a larger angle into the septal muscle. RBB damage could occur if the catheter tip is advanced anteroinferior from the distal HB or the tricuspid valve annulus, more than 20 mm toward the apex, not perpendicular to the septal muscle. Our visualization method,9 has demonstrated the following: (i) in HBP, both the HB and LBB bifurcation are usually implanted simultaneously in the thin interventricular muscle at the anteroinferior margin of the membranous septum; and (ii) the manualized proximal LBBP position10 does not cause RBB injury (Figure 1D and E). Perpendicular septal implantations of selective HBP exceeding 2 mm suggested that they penetrated the HB, covered by a thin septal muscle, to reach the initial LBB (Figure 1D and E). Conversely, a narrow QRS obtained by nonselective HBP or para Hisian pacing via a sub-valvular approach may indicate that the HB may not essentially be captured and the initial LBB is pacing in some cases. Compared with LBBP, HBP has a narrower target and is more difficult to capture and implant with an increased pacing threshold during follow-up,10 possibly hindering HBP use. Currently, the location of HBP capture remains unclear, considering the individual differences in the HB itself and the position of its branches; therefore, using terminologies such as selective or nonselective HB area pacing may be appropriate. On the other hand, CSP can easily be widely used on the broad target of LBBP, including for left bundle branch area pacing (LBBAP). However, it is less physiological pacing than HBP, and the stimulus is transmitted retrogradely to the RBB. The technique also visualizes the risk of acute septal perforation from deep implantation, especially when attempting to perform selective LBBP of the LBB just below the interventricular septal endocardium on the left ventricular side. The 3D navigation map emphasizes the difficulty in achieving both qualities of physiological pacing and technical capturability. Based on the in situ 3D CCS map, further progress in physiological pacing is expected. Overall, this new 3D navigation model not only improves understanding of the various outcomes of CSP but also contributes to the development of an accurate cardiac map. We sincerely thank those who donated their bodies to science so that current research could be performed with the approval of the Ethics Committee (Ref. #: A20001_A18015_A17033_A17005_25113_23011). The authors thank Mr. Makoto Sakai for his technical assistance. The data underlying this article will be shared upon reasonable request from the corresponding author. This study was supported by JSPS KAKENHI (no. 19K09278) and the Takeda Science Foundation.