Article ID: CJ-19-0399
Background: His-bundle pacing is an emerging routine technique that avoids pacing-dependent side effects. However, the success rate of His-bundle pacing is not 100%.
Methods and Results: Left bundle pacing or peri-left bundle pacing (LBP/peri-LBP) are recently developed techniques that directly capture the left bundle or ventricular tissue near the left bundle. We evaluated the success rate of LBP/peri-LBP in patients whose treatment with His-bundle pacing failed. In addition, we evaluated left ventricular contraction and desynchrony after LBP/peri-LBP.
Conclusions: LBP/peri-LBP is an alternative ventricular pacing method in atrioventricular block in patients with failure of His-bundle pacing.
His-bundle pacing (HBP) has become complementary to biventricular pacing, but the low success rate of HBP in patients with intrahisian block is a problem. Therefore, we hypothesized that left bundle pacing or peri-left bundle pacing (LBP/peri-LBP), which consists of stimulation of the conduction system below the His bundle (HB), might be feasible and effective in cases of HBP-implantation failure. Although a few case reports have presented the possibility of LBP/peri-LBP restoring electrical synchronization in patients with LBBB,1,2 in the present study we evaluated the LBP/peri-LBP success rate in patients in whom HBP had failed. This retrospective study was approved by the ethical committee of the Faculty of Medicine, the University of Tokyo (No. 2650-6), and written informed consents were given by all participants.
When HBP for atrioventricular block (AVB) failed, LBP/peri-LBP was applied. All 21 patients with greater than 2:1 AVB who needed permanent pacemaker implantation (female: 9 (43%); mean age: 76±3.2 years) had experienced failure of HBP and LBP had been attempted. Among them, LBP/peri-LBP succeeded in 17 cases. All patients with the intention to undergo LBP/peri-LBP had preserved left ventricular ejection fraction (LVEF) and mean LVEF was 65±1.7%. A total of 18 (86%) patients had infra-His block and 8 (38%) patients had bundle branch block (BBB) (right BBB: 4 (19%); complete left BBB (CLBBB: 4 (19%)). In patients with complete AVB, the heart rate of the escape rhythm was <30 beats/min.
The tip of the pacing lead (SelectSecureTM 3830, Medtronic) was first located at the HB’s electrogram-recorded site. The tip of this lead was then moved toward the right ventricular septum, together with the connecting line between the HB electrogram-recorded site and the right ventricular apex (the mean movement distance was 10–15 mm according to right anterior oblique (RAO) 25 degree). The lead was then placed just below the septal leaflet of the tricuspid valve (Figure A). The association between the tricuspid valve and the lead’s position was confirmed by contrast, which was infused by delivery catheter (C315TM, Medtronic). We confirmed that the lead for LBP/peri-LBP was not located on the septal leaflet of the tricuspid valve. The delivery catheter was also capable of providing backup support for advancing the tip of the lead by screwing-in to the left ventricular side. Before and during the process of advancing the lead to the left side, constant pacing (2 V/0.5 ms) was performed. Before the lead’s screwing-in, the QRS morphology by pacing showed LBB type (rS wave in V1) (Figure B, Middle panel). A small positive notch on the S wave in V1 was observed simultaneously if the lead was located in the appropriate position toward the LB. During the process of the lead’s screwing-in for the left side, this small positive notch on the S wave in V1 gradually increased its voltage, eventually becoming an R’ wave; ultimately, an rSR’ pattern in V1 (RBB type) morphology could be seen. In addition, during the rSR’ wave’s formation, the ventricular activation times in V4–6 were suddenly shortened (Figure B, Right panel). Once these ECG changes were observed, screwing-in was stopped to avoid exceeding the left ventricular endocardium. Thus, a narrow QRS complex with rSR’ wave in V1 was achieved (LBP/peri-LBP).
(A) Lead position for left bundle pacing (LBP)/peri-LBP. The pacing lead for LBP/peri-LBP is located between the temporary pacing lead located in the right ventricular apex and the His-bundle electrogram recorded site (the other pacing lead is temporarily located as a marker of the His-bundle pacing available site). (B) Morphological changes in pacing waves during screwing-in towards the left bundle from the right ventricular septum. The notch in V1 is the target morphology when the pacing lead for LBP/peri-LBP is located in an appropriate site in the right ventricular septum before screw-in (Middle bottom panel). Screwing-in for the left bundle is performed under continuous burst pacing to ensure its pacing morphology, which can reflect the depth of the lead within the ventricular septum. During screwing-in of the lead, this notch gradually moves upward and becomes an r’ wave and forms the rSr’ pattern (Right bottom panel). (C) Electrograms pre- and post-LBP (case 1) and peri-LBP (case 2). (D) Speckle-tracking echocardiography (bull’s eye plots display time to peak longitudinal strain) at the intrinsic and pacing beats during LBP. LBP did not provoke left ventricular dyssynchrony in the non-LBBB case (case 1). Peri-LBP recovered left ventricular dyssynchrony in the LBBB case (case 2). LAO, left anterior oblique; LBBB, left bundle branch; RAO, right anterior oblique.
HBP was successful in approximately 64% of cases of AVB, and the high ventricular pacing ratio predicted (≥40%) cases (n=92). However, in 21 patients, HBP was abandoned and treatment switched to LBP/peri-LBP because of the high pacing threshold of the HB or specific cases of non-selective HBP-like high pacing threshold of the ventricular septum even if the pacing threshold of the HB was acceptable. These strict judgments, whether HBP was safe for AVB cases or not, were performed to avoid loss of ventricular capture. Successful LBP/peri-LBP was achieved in 17 of 21 patients (success rate 81%), resulting in maintaining narrow QRS pacing (from 116±8.3 ms to 108±4.2 ms, P=0.41). In particular, a significant narrowing of the QRS complex was achieved in 4 patients with CLBBB among all patients with successful LBP/peri-LBP (n=17), from 151±4.0 ms to 122±6.7 ms (P=0.01).
Procedure time is an important factor in terms of new implantation methods. We perform the implantation of the HBP/LBP lead as follows. We first try to implant the HBP lead. The average implantation time of HBP lead is within 15 min in most of the successful cases. We do not spend more than 30 min trying to implant the HBP lead. More than this, we abandon HBP and immediately switch to implanting LBP/peri-LBP using the same lead. The procedure time of LBP implantation was within 15 min in all cases. As a result, the total implantation times for the LBP lead (<15 min) are shorter than those for the HBP lead (<30 min). There were no complications, including cardiovascular perforation, tricuspid valve injury, or loss of ventricular capture during the perioperative period.
Just after implantation, the mean LB/peri-LB capture threshold was 0.77±0.07 V/0.4 ms and the LBBB correction threshold in the CLBBB cases was 0.89±0.14 V/0.4 ms. At 6 months after LBP/peri-LBP lead implantation, mean LB/peri-LB capture thresholds and R wave amplitude had not deteriorated (pacing threshold from 0.75±0.07 V/0.4 ms (1 week after implantation) to 0.83±0.06 V/0.4 ms (6 months after implantation), R wave amplitude from 9.1±1.4 mV to 8.3±1.9 mV). The pacemaker checks were performed 1 week, 1 month and 6 months after implantation. There were also no complications, including cardiovascular perforation, tricuspid valve injury, loss of ventricular capture, or dislodgement of the leads, in the late phase. Therefore, the procedure of LBP/peri-LBP and long-term stability (≤6 months) might be acceptable even if the numbers in this study were small.
We evaluated LV synchrony under LBP/peri-LBP using speckle-tracking echocardiography in 2 representative patients with BBB (case 1: CRBBB; case 2: CLBBB). Case 1 was a 74-year-old female with 2:1 AVB in whom implantation of the HBP lead had failed because of its high pacing threshold and she was switched to LBP/peri-LBP. In patients with successful LBP/peri-LBP, including in this case, LB potential with an intrinsic rhythm could be observed in some of the patients without LBBB. In this case, LB potential was recorded from the lead tip during intrinsic rhythm (Figure C, Upper panel), and nonselective left bundle capture was achieved at this site. The interval from LB potential to ventricular activation (V wave) was shorter than the His potential–V interval. Case 2 was a 61-year-old male who presented with advanced AVB with CLBBB. His ECG waveform changed from a CLBBB pattern to a RBBB pattern after peri-LBP (Figure C, Lower panel). Even in patients in whom the LB potential could not be observed because of LBBB or no intrinsic junctional rhythm in AVB, the same narrow QRS complex was often established (we call this phenomenon “peri-LBP”).
Speckle-tracking echocardiography revealed no obvious deterioration of LV global longitudinal strain during LBP/peri-LBP compared with intrinsic rhythm (IR) (Case 1 IR: −23.1%, LBP/peri-LBP: −23.2%; Case 2 IR: −16.4%, LBP/peri-LBP: −15.4%). Impairment of LV synchrony by LBP/peri-LBP was not observed in case 1 (CRBBB) (Figure D). In CLBBB (case 2), recovery of activation delay in the anterior wall was observed by peri-LBP (Figure D).
Advisor and consultancy (K.F.) (Medtronic Japan), Endowed Chair (K.F.) from Medtronic Japan, Abbott Japan, Boston Scientific, Fukuda Densi and Biotronik Japan and research funding and scholarship funds (K.F.) from Medtronic Japan, Abbott Japan, Boston Scientific and Fukuda Densi.