From January 2021 to July 2021, a cohort of consecutive patients who met the criteria for elective permanent dual chamber pacemaker implantation, as outlined in the current recommendations (Class I) [9], were prospectively enrolled at the Beijing Anzhen Hospital. The inclusion criteria were: (1) age $>$18 years; (2) diagnosis of sick sinus syndrome (dual nodal lesions) and persistent high-grade atrioventricular block, with ventricular pacing ratio exceeding 70%; and (3) left ventricular ejection fraction (LVEF) $>$50% at baseline. Patients who met any of the following exclusion criteria were not included: presence of moderate to severe valvular disease; New York Heart Association (NYHA) functional class III or IV; history of acute and chronic coronary syndrome requiring intervention or revascularization within the previous 3 months; documented medical history of cardiomyopathy or tachyarrhythmia such as persistent atrial fibrillation; diagnosis of chronic obstructive pulmonary disease, liver failure, or kidney failure; current pregnancy; poor quality of ultrasound images; or inability to adhere to regular clinic follow-up appointments.

The included participants were assigned to either the LBBAP, RVSP, or RVAP group based on the site of pacing.

The research protocol was approved by the Beijing Anzhen Hospital Medical Ethics Committee (No. 2021083X) and was consistent with the ethical principles delineated in the 1975 Declaration of Helsinki. Patients provided written informed consent before study participation. The study was registered with the Chinese Clinical Trial Registry (No. ChiCTR2100048503).

As previously described [10], the LBBAP procedure was conducted with the Select Secure pacing lead (model 3830, 69 cm) delivered through the C315 fixed rigid curved sheath (Medtronic, Minneapolis, MN, USA). The LBB was initially assessed using a right anterior oblique 30^∘ fluoroscopic view. The appropriate pacing area was selected using the New Nine Partition Method [11]. The 3830 lead and C315 sheath were directed toward the interventricular septum and collectively spun in a clockwise direction. The lead was affixed by executing 8–10 clockwise rotations during the process of pace mapping. This was done after a W-shaped QRS morphology in V1 was seen at the tip of the lead. The lead was considered to have resolved when the QRS morphology in V1 exhibited a “QR/Qr” pattern, and the time interval from the pacing stimulus (Stim) to LV activation time (LVAT) in V5 or V6 was less than 75 ms and remained consistent regardless of high or low outputs. Threshold testing was conducted during the process of implantation using a Medtronic CareLink™ Programmer (Minneapolis, MN, USA). LBBAP was considered successful when all three of the following criteria were met, as stated by Zhang et al. [12]: (1) the final location of the 3830 lead, as determined by fluoroscopic imaging, was in the vicinity of the LBB; (2) unipolar-paced morphology exhibited a narrow QRS (130 ms) and an RBBB pattern; and (3) the time from Stim to peak LVAT in V4–V6 was 90 ms and remained consistently brief at both low and high outputs, indicating that the LBB was captured. Further indications of effective LBBAP were provided by direct LBB capture evidence, which included the LBB potential, selective LBBAP, and nonselective LBBAP detected by an electrophysiological recording device.

The RV leads were placed at either the RV apex or septum following standard procedure.

A 12-lead surface electrocardiogram (ECG) was performed at 100 mm/s with a GE Cardiolab Electrophysiology recording system (GE Healthcare Inc, Marlborough, MA, USA) before and after implantation. An intracardiac electrogram (IEGM) was obtained during implantation from the tip electrode of the 3830 lead. The intrinsic QRS duration (QRSd), paced QRSd, and Stim to LVAT time were measured sequentially. QRSd was calculated as the time from the onset to the end of the intrinsic or paced QRS complex in all 12 leads of the ECG. The Stim to LVAT time was measured as the interval from stimulus to the peak of the R wave in V4–V6.

Transthoracic two-dimensional (2D) echocardiography was conducted using S5-1 transducers and the EPIQ 7C ultrasound system (Philips, Bothell, WA, USA). The modified apical four-chamber view was utilized to acquire images of the RA and RV, with careful attention given to capturing the full structures of both. Current guidelines were applied to grade the severity of tricuspid regurgitation (TR), based on the vena contract width, proximal flow convergence, and regurgitant jet [13].

Measurement of the RV dimensions at the base during end diastole was conducted using an RV-focused apical 4-chamber view. RV systolic function is typically estimated by measuring the FAC, TAPSE, and the systolic velocity S-wave of the tricuspid annulus on tissue Doppler imaging (TDI). The calculation of the MPI, which represents a comprehensive measure of both systolic and diastolic RV performance, involves dividing the isovolumetric time intervals by the ventricular ejection time. The interventricular mechanical delay (IVMD) was employed to assess the mechanical synchronization between the two ventricles, as determined by pulse wave Doppler imaging. These parameters were obtained following the latest recommendations [14].

Analysis of 2D-STE data was conducted on 2D images of the RV-focused four-chamber view. The data were analyzed using the specialized software QLAB V.13.0 developed by Philips Medical Systems (Bothell, WA, USA). The data were acquired from four consecutive cardiac cycles. Concurrently with the cardiac cycles, an ECG was acquired to determine the heart rate. RA strain was applied during the reservoir phase (RASr), the conduit phase (RAScd), and the contraction phase (RASct). The RV strain parameters of RV free wall longitudinal strain (RVFWLS) and global longitudinal strain of the RV (GLSRV) were acquired according to the most recent recommendations [8]. Dysynchrony was evaluated as time-to-peak standard deviation (TPSD), and we also evaluated interventricular dysynchrony, based on left ventricular to right ventricular TPSD (LV-RV TPSD) using the same RR interval (The RR interval refers to the time interval or period between two consecutive R-waves in a cardiac cycle).

Patients’ clinical and demographic characteristics were recorded at baseline. During the stages of implantation, predischarge, and 1-year follow-up, data regarding the device were collected. Identical equipment was utilized for both the baseline and follow-up examinations. Two seasoned and impartial specialists, who were also blinded to the research, examined and analyzed the parameters.

The Shapiro-Wilk test was utilized to assess data normality. Continuous variables are presented as mean and standard deviation (SD) or median and interquartile range (IQR), whereas categorical variables are presented as frequency and percentage. Continuous variables were compared using the independent-samples t-test, and categorical variables were compared using the $\chi$² test. Data were compared among the three subgroups by one-way analysis of variance. Changes in variables from before to after follow-up in the same group were identified by paired t-test. Differences for which p 0.05 were deemed to be statistically significant. All statistical analyses were conducted using SPSS Statistics V22.0 (IBM Corp, Armonk, NY, USA).

A total of 197 consecutive patients underwent elective permanent dual-chamber pacemaker implantation during the study period. 97 were included in the study population after application of the inclusion and exclusion criteria, as outlined in Fig. 1. Another seven patients were then excluded from the final analysis due to unsatisfactory 2D-STE tracking data. One patient in the LBBAP group (1/31, 3.23% of total sample size) experienced an unsatisfactory outcome and was subsequently treated with RVSP. For the final analysis, each group included 30 patients.

Fig. 1.

Flow chart of the prospective cohort study. Abbreviations: UCG, ultrasound cardiography; STE, speckle tracking echocardiography; RT-3DE, real-time three-dimensional echocardiography; NYHA, New York Heart Association; MI, myocardial infarction; AF, atrial fibrillation; VP, ventricular pacing; LBBAP, left bundle branch area pacing; RVSP, right ventricular septal pacing; RVAP, right ventricular apical pacing.

Table 1 displays the demographic and clinical characteristics of the final study population. No statistically significant differences were observed in baseline characteristics, clinical disease diagnoses, underlying comorbidities, or medication use among the three groups.

The echocardiographic parameters for patients in each group are detailed in Table 2. No significant differences were detected among the three groups at baseline. At the 1-year follow-up, we observed that the LBBAP group could maintain right heart function, based on differences in RV FAC (50.81% $\pm$ 7.57% with LBBAP vs. 45.43% $\pm$ 6.91% with RVSP vs. 44.29% $\pm$ 7.98% with RVAP, p = 0.006), TAPSE (25.37 $\pm$ 4.7 mm vs. 22.83 $\pm$ 4.03 mm vs. 22.66 $\pm$ 3.96 mm, p = 0.017); and RVMPI (0.31% $\pm$ 0.15% vs. 0.37% $\pm$ 0.24% vs. 0.55% $\pm$ 0.33%, p = 0.015), while the RVSP and RVAP groups showed different degrees of impairment in right heart function. In addition to this, we found that no significant difference was seen in the change from baseline to follow-up in the LBBAP group for the following parameters: RVFAC (49.31 $\pm$ 7.6% with baseline vs. 50.81 $\pm$ 7.57% with follow-up, p = 0.679); TAPSE (23.43 $\pm$ 4.6 mm vs. 25.37 $\pm$ 4.7 mm, p = 0.122); RV MPI (0.45 $\pm$ 0.18% vs. 0.31 $\pm$ 0.15%, p = 0.93). RVSP resulted in a similar percentage reduction in TAPSE (p $>$ 0.05). In contrast, significant differences were observed in RVFAC (p = 0.09) and RVMPI (p = 0.037). RVSP resulted in a similar percentage reduction in TAPSE (p $>$ 0.05), while there were significant statistical differences in RVFAC (p = 0.009), RVMPI (p = 0.037). RVFAC, TAPSE, and RVMPI significantly decreased in the RVAP group. This indicates that LBBAP does not impair RV function in contrast to RVSP and RVAP.

At the 1-year follow-up, parameters of RV strain assessed by speckle tracking echocardiography (STE), including the global longitudinal strain of the RV (GLSRV) (–21.77% $\pm$ 5.99% with LBBAP vs. –18.82% $\pm$ 5.25% with RVSP vs. –17.33% $\pm$ 3.93% with RVAP, p = 0.001) and RV free wall longitudinal strain (RVFWLS) (–26.43% $\pm$ 5.53% vs. –22.71% $\pm$ 4.52% vs. –22.1% $\pm$ 5.92%, p = 0.008), were better in the LBBAP group than in the RVSP or RVAP group. Furthermore, there were no significant changes in GLSRV (from –21.62 $\pm$ 4.28% at baseline to –21.77 $\pm$ 5.99% at follow-up, p = 0.511) and RVFWLS (from –26.4 $\pm$ 6.05% to –26.43 $\pm$ 5.53%, p = 0.939) in the LBBAP group. Similar to the results in the LBBAP group, there were no significant differences from baseline to 1 year of follow-up in either GLSRV (from –20.17 $\pm$ 5.28% to –18.82 $\pm$ 5.25%, p = 0.203) or RVFWLS (from –24.24 $\pm$ 7.07% to –22.71 $\pm$ 4.52%, p = 0.353) in the RVSP group. In contrast, data from the RVAP group at the 1-year follow-up demonstrated a notable decline in both GLSRV (from –21.95 $\pm$ 6.29% at baseline to –17.33 $\pm$ 3.93% at follow-up, p = 0.029) and RVFWLS (from –26.92 $\pm$ 8.18% to –22.1 $\pm$5.92%, p = 0.025) (Table 3, Fig. 2).

Fig. 2.

RV strain curves after LBBAP, RVSP, or RVAP. Representative right ventricular (RV) strain curves for patients that underwent LBBAP (A), RVSP (B), or RVAP (C). In comparison to the other two groups, the LBBAP group exhibited the most uniform and consistent curves, while the RVSP group displayed a minor degree of disorder in its curves at the 1-year follow-up. Abbreviations: RV4CLS, right ventricular four-chamber longitudinal strain; RVFWLS, RV free wall longitudinal strain; LBBAP, left bundle branch area pacing; RVSP, right ventricular septal pacing; RVAP, right ventricular apical pacing.

Mechanical dysynchrony may be related to the severity of regurgitation. At the 1-year follow-up, patients who had undergone RVSP or RVAP had more severe TR compared to baseline, but this change in both TRVCM (0.33 $\pm$ 0.23 cm with baseline vs. 0.34 $\pm$ 0.11 cm with follow-up, p = 0.467) and TR jet area/right atrial area (RAA) (13.46 $\pm$ 1.33% vs. 13.34 $\pm$ 3.39%, p = 0.667) was not apparent in patients with LBBAP. Also at the 1-year follow-up, significant differences in the TR jet area/RAA (13.34% $\pm$ 3.39% vs. 15.44% $\pm$ 2.99% vs. 15.42% $\pm$ 3.90%, respectively; p = 0.015) and TR VCM (0.34 $\pm$ 0.11 cm vs. 0.40 $\pm$ 0.10 cm vs. 0.41 $\pm$ 0.12 cm, respectively; p = 0.035) were observed between patients who received LBBAP and those who received RVSP and RVAP (Table 2).

Among the groups, no statistically significant variation was identified in the structural or strain parameters of the RA, specifically the RA reservoir, conduit, and contraction strain, at either the initial assessment or over the 1-year follow-up period (Table 3).

As shown in Table 1, the QRS duration, which is regarded as a marker of electrical synchronization, was slightly increased in the LBBAP group (from 111.26 $\pm$ 24.07 ms to 114.45 $\pm$ 26.23 ms, p = 0.783), whereas QRS duration significantly increased in both RVSP (from 102.88 $\pm$ 18.78 ms to 133.5 $\pm$ 28.34 ms, p = 0.001) and RVAP (from 101.55 $\pm$ 16.09 ms to 157.29 $\pm$ 29.18 ms, p = 0.001). The LBBAP group showed a significant decrease in IVMD before and after pacemaker implantation (from 17.3 $\pm$ 26.5 to 8.6 $\pm$ 7.1 ms, p = 0.001), whereas IVMD increased significantly in both RVSP (from 17.6 $\pm$ 29.8 ms to 29.7 $\pm$ 9.9, p = 0.034) and RVAP (from 17.4 $\pm$ 28.4 to 44.7 $\pm$ 12.8, p = 0.001) (Table 2). Also, the LBBAP group showed a significant decrease in LV-RV TPSD during the 1-year follow-up [from 16.41 (8.81–42.5) to 12.28 (5.64–23.7), p = 0.001], while the LV-RV TPSD increased significantly during follow-up in both the RVSP group [from 18.91 (9.5–40.7) to 25.34 (8.81–53.6), p = 0.001] and RVAP group [from 16.47 (9.2–43.9) to 44.8 (19.6–147.5), p = 0.001] (Table 3).