Left bundle branch pacing compared to left ventricular septal myocardial pacing increases interventricular dyssynchrony but accelerates left ventricular lateral wall depolarization

Left bundle branch pacing compared to left ventricular septal myocardial pacing increases interventricular dyssynchrony but accelerates left ventricular lateral wall depolarization Karol Curila, MD, PhD, MSc,*Pavel Jurak, MSc, PhD, † Marek Jastrzebski, MD, PhD, ‡ Frits Prinzen, PhD, x Petr Waldauf, MD, PhD, { Josef Halamek, MSc, PhD, † Kevin Vernooy, MD, PhD, k Radovan Smisek, MSc, † ** Jakub Karch, MSc,* Filip Plesinger, MSc, PhD, † Pawel Moskal, MD, PhD, ‡ Marketa Susankova, MSc,* Lucie Znojilova, MSc,*Luuk Heckman, MD, k Ivo Viscor, MSc, PhD, † Vlastimil Vondra, PhD, † Pavel Leinveber, MSc, †† Pavel Osmancik, MD, PhD* From the *Cardiocenter, Third Faculty of Medicine, Charles University and University Hospital Kralovske Vinohrady, Prague, Czech Republic, † Institute of Scientific Instruments, the Czech Academy of Sciences, Brno, Czech Republic, ‡ First Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University, Medical College, Krakow, Poland, x Department of Physiology, Cardiovascular Research Institute Maastricht, Maastricht University Medical Center, Maastricht, the Netherlands, { Department of Anesthesia and Intensive Care, Charles University and University Hospital Kralovske Vinohrady, Prague, Czech Republic, k Department of Cardiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University Medical Center, Maastricht, the Netherlands, **Brno University of Technology, Faculty of Electrical Engineering and Communication, Department of Biomedical Engineering, Technick a 12, Brno, Czech Republic, and †† International Clinical Research Center, St. Anne’s University Hospital, Brno, Czech Republic. BACKGROUND Nonselective His-bundle pacing (nsHBp), nonselective left bundle branch pacing (nsLBBp), and left ventricular septal myocardial pacing (LVSP) are recognized as physiological pacing techniques. OBJECTIVE The purpose of this study was to compare differences in ventricular depolarization between these techniques using ultrahigh-frequency electrocardiography (UHF-ECG). METHODS In patients with bradycardia, nsHBp, nsLBBp (confirmed concomitant left bundle branch [LBB] and myocardial capture), and LVSP (pacing in left ventricular [LV] septal position without proven LBB capture) were performed. Timings of ventricular activations in precordial leads were displayed using UHF-ECG, and electrical dyssynchrony (e-DYS) was calculated as the difference between the first and last activation. Duration of local depolarization (Vd) was determined as width of the UHF-QRS complex at 50% of its amplitude. RESULTS In 68 patients, data were collected during nsLBBp (35), LVSP (96), and nsHBp (55). nsLBBp resulted in larger e-DYS than did LVSP and nsHBp [–24 ms (–28;–19) vs –12 ms (–16;–9) vs 10 ms (7;14), respectively; P,.001]. nsLBBp produced similar values of Vd in leads V 5 –V 8 (36-43 ms vs 38-43 ms; P5NS in all leads) but longer Vd in leads V 1 –V 4 (47–59 ms vs 41–44 ms; P,.05) as nsHBp. LVSP caused prolonged Vd in leads V 1 –V 8 compared to nsHBp and longer Vd in leads V 5 –V 8 compared to nsLBBp (44–51 ms vs 36–43 ms; P,.05) regardless of R-wave peak time in lead V 5 or QRS morphology in lead V 1 present during LVSP. CONCLUSION nslbbp preserves physiological LV depolarization but increases interventricular electrical dyssynchrony. LV lateral wall depolarization during LVSP is prolonged, but interventricular synchrony is preserved. KEYWORDS Depolarization duration; Dyssynchrony; His-bundle pacing; Left bundle branch pacing; Left ventricular septal myocardial pacing; Ultra-high-frequency electrocardiography (Heart Rhythm 2021;18:1281–1289) ©2021 Heart Rhythm Society. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Funding sources: This study was supported by the Charles University Research Program Q38, Research Centre Program No. UNCE/MED/002, 260530/SVV/ 2020; CAS Project RVO:68081731. Disclosures: Participating research institutions filed European patent application EP 19212534.2: “Method of Electrocardiographic Signal Processing and Apparatus for Performing The Method.”Address reprint requests and correspondence: Dr Karol Curila, Cardiocenter, Third Faculty of Medicine, Charles University, Srobarova 50, 100 34 Prague, Czech Republic. E-mail address: [email protected]. 1547-5271/© 2021 Heart Rhythm Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.hrthm.2021.04.025 Introduction In recent years, new pacing techniques such as His–Purkinje conduction system pacing have emerged. In addition to Hisbundle pacing (HBp), left ventricular septal myocardial pacing (LVSP) and direct capture of the left Tawara bundle (left bundle branch block pacing [LBBp]) using the transseptal approach have been described. 1,2 Both LVSP and LBBp are characterized by a pseudo–right bundle branch block (pseudo-RBBB) morphology in lead V 1 . They also share similar characteristics regarding lead placement in the left septum; however, LBBp is characterized by left septal subendocardial lead placement and confirmation of left bundle branch (LBB) capture using pacing maneuvers. 3 In contrast, left ventricular (LV) septal myocardial capture is achievable over a larger area and with shallower lead positions. 4 Ultra-high-frequency electrocardiography (UHF-ECG) analyzes the ultra-high-frequency components of ventricular myocyte action potentials. It shows the time sequence of ventricular depolarization and describes ventricular electrical dyssynchrony. It also displays local depolarization durations under individual chest leads, with shorter values indicating rapid conduction within specific ventricular segments. 5 Using UHF-ECG, we previously showed that nonselective His-bundle pacing (nsHBp) produces a physiological pattern of ventricular activation that is similar to intrinsic depolarization in patients with narrow QRSs. 6 The impact of LBBp and LVSP on electrical dyssynchrony and local depolarization durations, as seen using UHF-ECG, is not known. This work aimed to better understand the depolarization characteristics of both ventricles during LBBp and LVSP and to compare them to nsHBp, which was considered to represent physiological ventricular activation. Methods Study design and study population In this prospective study, consecutive patients with an indication for pacemaker implantation due to bradycardia were included in the Faculty Hospital Kralovske Vinohrady. The project was approved by the Ethics Committee, and all subjects signed informed consent before enrollment. Pacemaker implantation The left subclavian approach was preferred per study protocol. The His-bundle region was mapped using a SelectSecureÔlead (model 3830, 69 cm, Medtronic Inc., Minneapolis, MN), delivered through a fixed-curve sheath (C315HIS, Medtronic), and His-bundle signal was identified. Pacing in this location was performed to obtain nonselective His-bundle capture (nsHBp) as described previously. 7 The lead was then moved toward the right ventricle (RV), along a line between the His-bundle region and the RV apex, and screwed deep into the septum to obtain a position in the left side of the interventricular septum showing a paced QRS morphology of RBBB/pseudo-RBBB in lead V 1 .The process of screwing the lead into the interventricular septum was occasionally interrupted to check the paced QRS morphology and impedance. Once the paced QRS morphology (during unipolar pacing with 5V output at 0.5 ms) demonstrated a terminal r/R or rs/Rs morphology in lead V 1 , the pacing output was decreased until capture was lost. During decreased pacing output, QRS morphology and changes in R-wave peak time (RWPT) in V 5 , which represents the time between the pacing artifact and the moment the electrical wavefront reaches the epicardium under lead V 5 , 8 were checked and recorded. Two distinct capture types in the left septal area were described (Figure 1): Table 1 Patient characteristics Age (y) 78 69 Male 42 (61) Comorbidities Heart failure 14 (21) Coronary heart disease 27 (40) Diabetes mellitus 32 (47) Hypertension 54 (79) LV ejection fraction (%) 56 67 Septal thickness 11 62 Pacing indications AV block 39 (57) SSS 20 (29) Bifascicular/trifascicular block 10 (15) Atrial fibrillation with planned AV junctional ablation 3 (4) QRS morphology LBBB 9 (13) RBBB 21 (31) Isolated left Tawara fascicular blocks 3 (4) Left Tawara fascicular blocks with RBBB 15 (22) NIVCD 3 (4) Narrow QRS 32 (47) Values are given as mean 6SD or n (%). AV 5atrioventricular; LBBB 5left bundle branch block; LV 5left ventricle; NIVCD 5nonspecific intraventricular conduction delay; RBBB 5 right bundle branch block; SSS 5sick sinus syndrome. Table 2 Parameters measured during each type of ventricular capture Type of capture V 5 RWPT (ms) Lead depth (mm) LBBpo/HBpo present N (%) Actual QRSd (ms) nsLBBp (n 5 35) 70 (66;73)*14 (13;15) † 28 (80) ‡ 104 (100;108) LVSP (n 596) 86 (84;89) 11 (11;12) 7 (7) 103 (100;107) nsHBp (n 555) 88 (86;91) NA 55 (100) 93 (89;97) x HBpo 5His-bundle potential; LBBpo 5left bundle branch potential; LVSP 5left ventricular septal myocardial pacing; nsHBp 5nonselective His-bundle pacing; nsLBBp 5nonselective left bundle branch pacing; QRSd 5QRS duration; RWPT 5R-wave peak time. *P,.01 vs LVSP and nsHBp. † P,.001 vs LVSP. ‡ P,.001 vs LVSP. x P,.001 vs nsLBBp and LVSP. 1282 Heart Rhythm, Vol 18, No 8, August 2021 1. Nonselective left bundle branch capture (nsLBBp), that is, concomitant myocardial and LBB capture, defined by a pseudo-RBBB morphology with the terminal r/R in lead V 1 , which occurred after increasing the pacing output up to 5 V at 0.5 ms from selective left bundle branch capture (sLBBp) or myocardial LV septal capture. During the transition from sLBBp, changes in QRS complex and EGM morphology (narrowing of Rʹin lead V 1 and disappearance of a discrete LBB electrogram) were seen; however, V 5 RWPT remained the same. During the transition from myocardial capture to nsLBBp, V 5 RWPT shortened .10 ms and, usually, the amplitude of R in lead V 1 increased or changed from rs/Rs to r/R morphology. 9 2. Left ventricular septal myocardial capture (LVSP), that is, pure myocardial capture of the left septum in the location where LBB capture was not observed with output up to 5 V at 0.5 ms, was defined by a narrower paced QRS (compared to pacing from the RV septum) and a terminal R/r or rs/Rs morphology in lead V 1 . While decreasing the pacing output from 5 V to loss of capture, either no changes or only minor changes in QRS morphology and V 5 RWPT (,10 ms) were observed. We aimed to fix the lead in a location providing nsLBBp with a normal QRS axis and the transition to sLBBp or myocardial pacing during each lead deployment. If the "drilling effect” 10 or impedance values close to 500 Uprohibited deeper lead deployment, the lead was withdrawn and deployed in a different location. If nsLBBp was not observed in the second tested location, no further attempts were made. UHF-ECG data acquisition and analysis of other measured parameters A ventricular dyssynchrony imaging (VDI) monitor (ISI Brno, Cardion, FNUSA, Czech Republic) was used to record and analyze the 5-kHz, 14-lead ECG signals with 3-nV resolution and a frequency range of 1.5 kHz. Standard V 1 -V 8 chest lead positions were used, except for lead V 1 , which was moved from the 4th to 5th right parasternal intercostal space to obtain better signals from the lateral RV wall. UHF-ECG data for all captures were collected during 2–3 minutes of VVI pacing at 110 bpm. Signal processing and UHFECG map construction were described previously. 5 In brief, median amplitude envelopes were computed for 16 Figure 1 A, B: Left septal lead placement with nsLBBp to left septal myocardial capture transition (A) and nsLBBp to sLBBp (B). C: LVSP with paced terminal “R”morphology in V 1 , without a significant change of QRS morphology and V 5 RWPT between pacing at 5 V and above the capture threshold. D: LVSP with paced terminal “rs”morphology in V 1 , without a significant change of QRS morphology and V 5 RWPT between pacing at 5 V and above the capture threshold. LA 5left atrium; LV 5left ventricle; LVSP 5left ventricular septal myocardial pacing; nsLBBp 5nonselective left bundle branch pacing; RV 5right ventricle; RWPT 5R-wave peak time; sLBBp 5selective left bundle branch pacing. Curila et al UHF-ECG Study of Physiological Pacing 1283 frequency bands (150–1000 Hz) for each chest lead. The broad-band QRS complex (UHF-QRS) was constructed as the average of the 16 normalized median amplitude envelopes and displayed as a color map for each lead. Local activation times were calculated as the center of mass (Mxc) of UHF-QRS above 50% threshold of the baseline-to-peak amplitude in each chest lead. Local depolarization duration under leads V 1 –V 8 was computed as UHF-QRS duration at 50% of its amplitude (Vxd parameter) (Figure 2A). e-DYS, a parameter of interventricular dyssynchrony (maximum difference between M1-8c), and Vd mean (mean value of V1-8d) were calculated (Figure 2C). Positive e-DYS indicates delayed LV activation, and negative e-DYS indicates delayed RV activation. The height of the lead tip on the septum in the right anterior oblique (RAO) projection was measured with respect to heart silhouette diameter and reported in percentages as reported recently (Supplemental Figure 1). 11 Global QRS duration (QRSd) values were measured using an electrophysiological system (LabSystem Pro, Boston Scientific, USA), from the earliest to the last deflection in any of the 12 leads. During pacing, the beginning of the QRS was measured from the pacing artifact (QRSd) and from the first deflection identified after the pacing artifact (actual QRSd). Paced V 5 RWPT was measured from the pacing artifact to the maximum positive QRS amplitude in lead V 5 . All measurements were obtained at 200 mm/s using 2 consecutive beats, and their average values were taken. During the procedure, 2–3 mL of contrast agent was injected through a C315HIS sheath in the left anterior oblique (LAO) projection. Lead depth inside the septum for all pacing locations was measured with an xViewer (Vidis, Prague, Czech Republic) using the distance between the tip and the anodal ring of the 3830 lead (10.8 mm) in LAO as a reference. QRS axis in the frontal plane was calculated and considered to have deviated to the left if axis 230, normal (axis 229to 105), or deviated to the right (axis 105). Figure 2 Presentation of UHF-QRS complexes, Mxc activation times, and Vxd local depolarization duration calculation for a patient with left anterior hemiblock and right bundle branch block. A: Ultra-high-frequency amplitude envelopes of QRS complexes (UHF-QRS), Mxc computed as the center of mass above 50% threshold of the baseline-to-peak amplitude, and Vxd determined as depolarization duration at 50% threshold of the baseline-to-peak amplitude in V6 and V3 (V6d and V3d). B: Twelve-lead electrocardiogram (ECG). C: Ventricular depolarization map with visualization of M1-8c, electrical interventricular dyssynchrony (e-DYS), and V1-8d values. (For details, see Jurak et al. 5 )InC, the dark line connects the center of masses (solid points) under the specific lead (displayed on the y-axis). Time (ms) is displayed on the x-axis. In this case, the first activation occurred under V7 (M7c) and the last under V 1 (e-DYS 5257 ms). The width of depolarization under V3 is indicated by the green arrow (V3d) and under V6 by the yellow arrow (V6d). The numerical parameters of the local depolarization duration (under each lead) are shown on the right. 1284 Heart Rhythm, Vol 18, No 8, August 2021 Statistical analysis Exploratory data analysis was performed for all parameters. Unpaired comparisons of continuous and categorical variables were made using the unpaired Student ttest and c 2 test. Repeated measurement comparisons were made using a linear mixed effect model and the Tukey multiple comparison test. The results of these models are given as mean (95% confidence interval) and comparisons as mean differences (95% confidence interval) and Pvalues (Figures 3–6). P ,.05 was considered significant. RStudio Version 1.2.1335 with R Version 3.6.1 was used to perform statistical analyses. The linear mixed effect model was calculated using lme4 Version 1.1–21. If not specified, all values are given as mean (95% confidence interval). Results In a total of 68 patients with bradycardia, we recorded UHFECG during nsLBBp (n 535), LVSP (n 596), and nsHBp (n 555). Narrow QRS with duration 120 ms was present during spontaneous rhythm in 47% of patients. In 32 patients, .1 LVSP was investigated. Patient characteristics are listed in Table 1. The shortest V 5 RWPT was observed during nsLBBp, followed by LVSP and nsHBp. During nsLBBp, the lead tip was deeper in the septum than during LVSP, and the LBB potential was present in the majority of nsLBBp (28/35) compared to only 7 of 96 LVSP (Table 2). All LVSP with a visible LBB potential had paced V 5 RWPTs that were .10 ms longer than their LBBpotential RWPTs in V 5 during spontaneous rhythm, which indicated that LBB was not captured (Supplemental Figure 2). A comparison of the sequence of ventricular activation between nsLBBp, LVSP, and nsHBp is shown in Figure 3A. The timing of ventricular depolarization under leads V 4 –V 8 was remarkably similar during both nsLBBp and LVSP. However, more delayed activation under leads V 1 –V 3 during nsLBBp resulted in greater left-to-right interventricular dyssynchrony with a more negative e-DYS than during LVSP (Figure 3B). Compared to both left septal captures, mean eDYS during nsHBp was 10 ms (7;14) (P,.001). Negative e-DYS (indicating delayed RV depolarization) was observed in 7 of 55 nsHBp (in 5 due to incomplete RBBB correction), 34 of 35 nsLBBp, and 75 of 96 LVSP. nsHBp resulted in the shortest QRSd and Vd mean (Figures 4Aand4B). There was no difference in QRSd and Vd mean between nsLBBp and LVSP, but significant differences in Vd for V 1 –V 8 were observed. Local depolarization durations associated with the LV lateral wall (V5d– V8d) were shorter; however, depolarization duration in V 1 –V 3 was longer during nsLBBp than LVSP. V5d–V8d values observed in nsLBBp were statistically the same as those in nsHBp, but both nsLBBp and LVSP had longer V1–4d compared to nsHBp (Figure 4C). V 5 RWPT values overlapped during LVSP (range 52–114 ms) and nsLBBp (52–88 ms), with 51 of the LVSP (53%) having V 5 RWPT 88 ms. To determine whether shorter V 5 RWPT values during LVSP reflect faster LV lateral wall depolarization, we compared LVSP with the shortest and longest V 5 RWPTs divided by quartiles. Mean V 5 RWPT was 69 ms in the 1st quartile vs 101 ms in the 4th quartile. We found that only 2 of 24 captures from the 1st quartile had a normal axis compared to 17 of 24 from the 4th quartile (P,.001). Those from the 1st quartile had more significant interventricular dyssynchrony [e-DYS 220 ms (227;-14) vs 28ms(214;22); P5.007], but individual V1-8d values were the same (Figure 5). LVSP with left-axis deviation (51/96) had shorter V 5 RWPT of 80 ms (77;83) compared to those with normal axis of 92 ms (89;95) (P,.001). To study whether LVSP LVSP LVSP LVSP *** *** *** AB Figure 3 Local activation times M1c-8c (first activated segment was placed on 0 ms) (A) and comparison of e-DYS between nsLBBp, LVSP, and nsHBp (B). CI 5confidence interval; nsHBP 5nonselective His-bundle pacing; other abbreviations as in Figures 1 and 2. ***P,.001. Curila et al UHF-ECG Study of Physiological Pacing 1285 with left-axis deviation were placed more inferiorly on the septum, such as those with a normal axis, we calculated the height of the lead tip in the RAO projection with respect to the height of the heart silhouette. Stored RAO projections were available in 17 LVSP with left-axis deviation and 14 LVSP with normal QRS axis. Lead tip positions resulting in LVSP with left-axis deviation were in RAO placed lower than those with normal axis [42% vs 50%; mean difference –8% (–13;–3); P5.004] and had shorter V 5 RWPT [81 ms vs 91 ms; mean difference –10 ms (–18;–2); P5.02]. To understand how V 1 QRS morphologies influence ventricular depolarization patterns, we compared 75 LVSP having terminal r/R and 21 LVSP with terminal rs/Rs morphology in V 1 . Captures with terminal rs/Rs morphology had more shallow lead tip insertions [10 mm (8;11) vs 12 mm (11;12); P5.01] and less interventricular dyssynchrony [e-DYS 23ms(29;4) vs 216 ms (219;213); P,.001] compared to those with terminal r/R morphology. However, their Vd in V2–V8 was the same (Figure 6). Eleven of 21 captures with terminal rs/RS morphology had a positive e-DYS compared to 10 of 75 captures with terminal r/R morphology (P,.001). No differences in e-DYS, Vd mean , QRSd, and local V18d were observed when comparing left septal pacing with left-axis deviation (n 566) to those with normal paced QRS axis (n 562), although V5 RWPT was significantly shorter with left-axis deviation [76 ms (72;80) vs 85 ms (82;89); P,.001]. The proportions of nsLBBp and LVSP in both groups were the same: 27% nsLBBp and 73% LVSP in the left-axis deviation group vs 26% and 74%, respectively, in the normal axis group (P5.99) (Supplemental Figure 3). ** ** *** *** LVSP LVSP LVSP LVSP A C B Figure 4 QRS duration (QRSd) (A),Vd mean (B), and local depolarization durations (Vd in V 1 –V 8 )(C) between nsLBBp, LVSP, and nsHBp. Abbreviations as in Figures 1 and 2.**P,.01, ***P,.001. 1286 Heart Rhythm, Vol 18, No 8, August 2021 Discussion This study showed that, in bradycardia patients, HBP provides more physiological activation than pacing from the left side of the interventricular septum, and that differences in UHF-ECG–derived ventricular synchrony and local depolarization duration during pacing from the left side of the interventricular septum depend on LBB capture. nsLBBP creates greater interventricular dyssynchrony compared to LVSP. LV lateral wall depolarization durations in nsLBBp are similar to those in nsHBP due to direct capture of His– Purkinje tissue with rapid electrical wavefront propagation. The study also showed that left septal myocardial pacing without confirmed LBB capture (LVSP) has slower LV lateral wall depolarization compared to nsLBBp and nsHBp irrespective of RWPT in V 5 or V 1 QRS morphology. Finally, both nsLBBp and LVSP compared to nsHBp resulted in longer depolarization durations in leads placed above the septum and the RV. This likely is a result of the greater contribution of slower myocardial conduction in their activation. In past years, novel pacing strategies have been introduced to avoid nonphysiological ventricular activation associated with RV myocardial pacing. Shortcomings linked to HBP led to the development of transseptal pacing methods. 1,2 Both nsLBBp and LVSP reduce the transseptal conduction time associated with RV myocardial pacing. They also share other similarities, such as the lead tip being fixed in the left septal area, and delayed activation of the RV leading to a typical QRS morphology with the terminal r/R in lead V 1 . 3 However, direct capture of the LV subendocardial His– Purkinje system during LBBp results in earlier electrical wavefront propagation through the LV cavity compared to myocardial capture of the LV septum. This is reflected by a shorter V 5 RWPT compared to pure myocardial LV septal capture, in which the left septal myocytes are activated first after pacing and then, after some delay, the LV His– Purkinje system is activated. Due to the ability to display local depolarization durations using UHF-ECG, we were able to show that LVSP caused statistically significant prolongation of local depolarization durations under V 5 –V 8 compared to nsLBBp. This likely results from the more significant contribution of slower myocardial cell-to-cell electrical wavefront propagation through the LV during LVSP. However, V5d–V8d values during nsLBBp were the same as during nsHBp, which shows that the latency between the pacing artifact and LBB capture in LBBp does not play a significant role in LV lateral wall depolarization. The interventricular dyssynchrony (expressed as e-DYS) was significantly shorter during LVSP than during nsLBBp. Better RV–LV depolarization interplay during LVSP was a result of the time delay between left septal myocyte capture and activation of the LV His–Purkinje system. The longer the delay, the less interventricular dyssynchrony is present because left-to-right transseptal depolarization occurs Figure 5 Local depolarization durations (Vd in V 1 –V 8 ) during LVSP with the shortest and longest RWPT in V 5 (1st quartile vs 4th quartile). Abbreviations as in Figures 1 and 2. Curila et al UHF-ECG Study of Physiological Pacing 1287 immediately after pacing; however, LV conduction system activation with the fastest LV depolarization is delayed. Criteria to differentiate between LBBp and left septal myocardial pacing are still not established, although a few attempts have been made. 3,12 Lead tip penetration into the septum leads to distinct paced QRS morphology changes in V 1 . The notch, present in the nadir location in the QRS complex during right septal pacing, moves upward as the lead progresses into the septum. 13 Finally, formation of the pseudo-RBBB in V 1 usually is considered a sign that the left septal area was reached. However, it should not be considered a marker of optimal LV depolarization because it just reflects delayed RV depolarization compared to the relatively faster depolarization of the LV. As we showed in our study, pseudo-RBBB morphology in V 1 was also present in more shallow LV septal positions, where direct LBB capture was not proven with pacing output up to 5 V at 0.5 ms. Although they resulted in improved interventricular synchrony, their LV lateral wall depolarization durations were longer compared to nsLBBp. Another parameter commonly used during left septal lead placement is V 5 RWPT. It represents the time between the pacing artifact and the moment the electrical wavefront reaches the LV lateral wall epicardium, 8 and its shorter values could assume faster LV activation. However, as we showed, shorter V 5 RWPTs during LVSP were not a reflection of faster LV lateral wall depolarization but rather were the result of lead deployment closer to the left posterior LBB fascicle with resultant left-axis deviation. Lead placement in more inferior septal areas with preferential activation of the His–Purkinje tissue of the left posterior fascicle shorten V 5 RWPT because the distance to the LV lateral wall epicardium is shorter compared to pacing the LBB trunk area in more superior locations. The most balanced LV–RV depolarization was seen during LVSP with the terminal rs/Rs morphology in V 1 due to reduced transseptal left-to-right conduction time because of more shallow lead placement compared to deeper LVSP lead positions with r/R V 1 morphology. It not only resulted in less interventricular dyssynchrony but also produced shorter depolarization durations under V 1 . Compared to nsHBp, both types of left septal pacing led to significant prolongations of depolarization under the leads located above the interventricular septum and RV. This is caused by leftto-right electrical wavefront propagation through the septum during nsLBBp and LVSP, and delayed RV activation due to slowed conduction associated with myocardial cell-to-cell transmission. Both the septum and the RV are depolarized in an unphysiological manner during nsLBBp and LVSP, Figure 6 Local depolarization durations (Vd in V 1 –V 8 ) during LVSP and with terminal r/R morphology and rs/Rs morphology in V 1 . Abbreviations as in Figures 1 and 2. 1288 Heart Rhythm, Vol 18, No 8, August 2021 which is in contrast to the physiological activation seen during nsHBp. Study limitations This study was performed during the actual implant procedures. UHF-ECG measurements were taken immediately after the lead was placed in a predefined position and ventricular capture type was confirmed. We cannot rule out that the resultant damage to conductive and myocardial tissue could have influenced the paced ventricular depolarization patterns. Data were not compared to that of any other invasive or noninvasive electrocardiographic methods, and no hemodynamic or echocardiographic measurement of mechanical dyssynchrony during the procedure was performed. The only method used to confirm LBB capture was the decrease in pacing amplitude, which could have led to misclassification of some nsLBB captures as LVSP captures. The low QRS signal quality did not allow construction of UHF-ECG maps in 1 patient, and this patient was excluded from the study. Because of the low number of patients with LBBB during spontaneous rhythm included in the study, the differences in ventricular depolarization between nsLBBp and LVSP cannot be generalized to the population of patients with LBBB. Conclusion This UHF-ECG study comparing left septal pacing with HBP showed that both types of left septal pacing cause less physiological ventricular depolarization compared to nsHBp. Nonselective capture of the left bundle preserves fast LV lateral wall depolarization but at the costs of deterioration in interventricular synchrony and slower conduction in the septum and RV. LV septal lead placement, without proven LBB capture, produces less interventricular dyssynchrony than nsLBBp. However, the contribution of rapid conduction in all individual ventricular segments is smaller compared to the physiological ventricular activation seen during nsHBp. If the most physiological pattern of LV activation is the goal of the implant procedure, then nsHBp and nsLBBp should be preferred over LVSP because LVSP results in slower LV lateral wall depolarization, irrespective of V 5 RWPT or paced QRS morphology in lead V 1 . Appendix Supplementary data Supplementary data associated with this article can be found in the online version at https://doi.org/10.1016/j.hrthm.2021. 04.025. References 1. Huang W, Su L, Wu S, et al. A novel pacing strategy with low and stable output: pacing the left bundle branch immediately beyond the conduction block. Can J Cardiol 2017;33. 1736e1731-1736–e1733. 2. Mafi-Rad M, Luermans JG, Blaauw Y, et al. Feasibility and acute hemodynamic effect of left ventricular septal pacing by transvenous approach through the interventricular septum. Circ Arrhythm Electrophysiol 2016;9:e003344. 3. Huang W, Chen X, Su L, et al. A beginner’s guide to permanent left bundle branch pacing. Heart Rhythm 2019;16:1791–1796. 4. Wu S, Sharma PS, Huang W. Novel left ventricular cardiac synchronization: left ventricular septal pacing or left bundle branch pacing? Europace 2020;22(Suppl 2):ii10–ii18. 5. Jurak P, Curila K, Leinveber P, et al. Novel ultra-high-frequency electrocardiogram tool for the description of the ventricular depolarization pattern before and during cardiac resynchronization. J Cardiovasc Electrophysiol 2020;31:300–307. 6. Curila K, Prochazkova R, Jurak P, et al. Both selective and nonselective His bundle, but not myocardial, pacing preserve ventricular electrical synchrony assessed by ultra-high-frequency ECG. Heart Rhythm 2020;17:607–614. 7. Vijayaraman P, Dandamudi G, Zanon F, et al. Permanent His bundle pacing: recommendations from a Multicenter His Bundle Pacing Collaborative Working Group for standardization of definitions, implant measurements, and follow-up. Heart Rhythm 2018;15:460–468. 8. Perez-Riera AR, de Abreu LC, Barbosa-Barros R, et al. R-peak time: an electrocardiographic parameter with multiple clinical applications. Ann Noninvasive Electrocardiol 2016;21:10–19. 9. Su L, Wang S, Wu S, et al. Long-term safety and feasibility of left bundle branch pacing in a large single-center study. Circ Arrhythm Electrophysiol 2021;14:e009261. 10. Jastrzebski M, Moskal P, Holda MK, et al. Deep septal deployment of a thin, lumenless pacing lead: a translational cadaver simulation study. Europace 2020;22:156–161. 11. Curila K, Jurak P, Halamek J, et al. Ventricular activation pattern assessment during right ventricular pacing: ultrahigh-frequency ECG study. J Cardiovasc Electrophysiology 2021 Mar 8; https://doi.org/10.1111/jce.14985. 12. Jastrzebski M, Kielbasa G, Curila K, et al. Physiology-based electrocardiographic criteria for left bundle branch capture. Heart Rhythm 2021 Mar 4; https://doi.org/ 10.1016/j.hrthm.2021.02.021. 13. Jastrzebski M, Moskal P, Bednarek A, et al. Programmed deep septal stimulation: A novel maneuver for the diagnosis of left bundle branch capture during permanent pacing. J Cardiovasc Electrophysiol 2020;31:485–493. Curila et al UHF-ECG Study of Physiological Pacing 1289