Cardiac Resynchronization Therapy Using Left- bundle-branch Area Pacing and Coronary Sinus Pacing


 Background: Cardiac resynchronization therapy via biventricular pacing (BVP) is an established therapy for patients with heart failure. Recently, it has been shown that left bundle branch area pacing (LBBAP) is feasible and may also improve clinical outcomes. In this article, we describe a new technique (sequential LBBAP followed by coronary sinus pacing, designated LOT-CRT) and assess the feasibility of LOT-CRT.Methods: The database of all patients with adaptive CRT from single centre was reviewed retrospectively. The eligible patients were divided into two groups, LOT-CRT and BV-CRT. The LBBAP lead implanted using our methods. The QRS duration (QRSd) was measured at baseline and during LBBAP, BVP, and LOT-CRT.Results: The study enrolled 17 consecutive heart failure patients with LBBB. LBBAP failed in 1 patient, succeed in 8 patients, while CS leads were implanted successfully in all patients. At baseline, the two groups (8 cases in LOT-CRT group, 9 cases in BV-CRT group) were matched for QRSd and ischemic cardiomyopathy (ICM, 5 cases in LOT-CRT group, 4 cases in BV-CRT group). In LOT-CRT group, BVP resulted in significant reduction of the QRSd from 158.0 ± 13.0 ms at baseline to 132.0 ± 4.5 ms (P=0.019). Compared with BVP, unipolar LBBAP resulted in further reduction of the QRSd to 123.0 ± 5.7 ms (P < 0.01). However, LOT-CRT with adaptive algorithm resulted in a significantly greater reduction of the QRSd to 117.0 ± 6.7 ms (P < 0.01). In BV-CRT group, BVP resulted in significant reduction of the QRSd from 176.7 ±19.7 ms at baseline to 143.3 ±8.2 ms (P=0.011). However, compared with LOT-CRT, BVP has no any advantage in reducing QRSd (P >0.05). As compared to the baseline after 3 months of LBBAP, patients in LOT-CRT group showed significant improvement in LVEF and NT-proBNP levels (P < 0.01), while patients in BV-CRT group showed non-significant changes in these parameters (P >0.05).Conclusions: The study demonstrates that LOT-CRT is clinically feasible in patients with systolic HF and LBBB. LOT-CRT was associated with significant narrowing of the QRSd and improvement in LV function, especially in patients with ICM.

(Group 1), conventional CRT using biventricular pacing (BV-CRT, Group2). If primary LBBAP was unsuccessful in patients for LOT-CRT, BIV-CRT would be implanted and the patient was transferred into group 2. Correspondingly if primary CS lead was unsuccessful in patients with indication for CRT, a LBBAP lead was implanted and the patient could be exited from the study.

Procedural Details
The RV de brillator lead was rst implanted in the RV to provide backup ventricular pacing should the patient develop transient complete atrioventricular block during LBBAP lead placement. Subsequently, the coronary sinus (CS) lead was implanted using routine implantation techniques, targeting sites with maximal LV delay [11,12]. Then, LBBAP was performed using the Select Secure pacing lead. All de brillator electrodes were implanted in the RV apical position. The uoroscopy durations for the entire procedure, LBBAP lead implantation and LV lead implantation were separately recorded.

LBBAP lead implantation technique
As previously described [13][14][15][16], a Select Site C315 His sheath and a Select Secure 3830 pacing lead (Medtronic Inc, Minneapolis, MN, USA) were advanced to the implantation site. The right ventricular septal location for LBBAP was identi ed using the anatomical location and pacing localization the ninegrid system [17]. Once this site was identi ed, the pacing lead is advanced deep into the septum while the unipolar pacing impedance, electrogram characteristics and paced QRS morphology were monitored.
Additionally, the lead orientation can be displayed in various projections. During the initial LBBAP lead xation, if the lead twists back, this indicates that the lead and sheath are not oriented orthogonal to the RV septum. Generally, the sheath and the lead are oriented such that the lead is pointing in the 12-to 1o'clock direction from a right anterior oblique viewing angle of 30° and the 2-to 3-o'clock direction from a left anterior oblique viewing angle of 30°[18].
If an acceptable LBB capture could not be achieved after 5 attempts of lead positioning, it was considered a failure [19].

Optimal CS location
The details of the device and procedure have been described elsewhere [11,12]. Optimal vein selection and lead implantation is greatly facilitated by high-quality occlusive venography. Traditionally, CS intubation is performed by advancing a 0.035-inch hydrophilic wire to the region of the CS ostium via a preformed guide catheter and probing to locate the CS ostium. Venograms are typically performed in the anteroposterior and left anterior oblique projections. Optimal CS location was limited to the distribution of the coronary veins [11,12].

Device Connection and Programming
In group 1, the patients undergoing CRT-de brillator treatment, the LBBAP lead was connected to the pace-sensing portion of the RV port, and a CS lead was connected to the LV port. The pace-sensing portion of the spliced implantable cardioverter -de brillator (ICD) lead (DF-1) was capped. In patients undergoing CRT-pacemaker (P) treatment, the CS lead was connected to the LV port. Then the LBBAP lead was connected to the RV port.
In group 2, the patients undergoing CRT-de brillator treatment, the CS lead was connected to the LV port.
Then the RV de brillator lead was connected to the RV port.

Implant Measurements
The pacing output required to maximally narrow the QRS (BBB correction threshold) and LBB capture threshold (without BBB correction) was assessed and recorded at a pulse width of 1.0 ms. The QRSd values at baseline and during LBBAP, BVP (over RV de brillator lead and the CS lead during procedure period rstly, or via the lead at the LBBA when RV de brillator lead unavailable) and LBB-optimized LV pacing were measured on the EP recording system at 100 mm/s. The stimulus to left ventricular activation time during LBBAP was documented.
Programming and follow up Before hospital discharge, separate "zones" can be programmed for detection of ventricular brillation and ventricular tachycardia. All patients were seen for routine clinical follow-up at standard time intervals (every 3 months) and had a follow-up period of at least 3 months. Functional status was assessed by the NYHA classi cation system. Device thresholds were checked and adjusted as needed to maximize battery longevity. The pacing threshold, impedance and R wave amplitude were measured. All devicedetected and treated VT/VF episodes were reviewed and adjudicated by an independent episode reviewer.
According to previous literature[20], a high pacing threshold was de ned as a pacing threshold over 2.5 V/0.4 ms or an increase of more than 1.0 V compared with the baseline after the procedure and at followup. Echocardiographic indices, including LVEF, LV end-diastolic dimension (LVEDD), and pulmonary artery systolic pressure, were recorded before implantation and at follow-up.

Statistical analysis
Continuous variables are presented as the mean ± SD or median. Paired comparisons were made using Student's t-test if the data were normally distributed; otherwise, the nonparametric Wilcoxon signed-rank test was used. Paired categorical data (NYHA functional class) were compared using the Wilcoxon test. P ≦ 0.05 was considered signi cant.

Results
Four out of the 16 eligible patients were excluded from the study according to the exclusion criteria. Consequently, 12 patients were eligible for CRT. During follow-up, one patient in group 2 lost to follow-up.
So out of the 12 eligible patients, only 11 patients enrolled the study (6 in group 1, 5 in group 2). Among 6 patients in group1, 1 patient failed to complete LBBAP and transferred into group 2. In the end, ve patients (45.5%) were de ned as group LOT-CRT; six patients (54.5%) were de ned as group BIV-CRT.
All patients had had at least 1 HF hospitalization 3 months before LBBAP implantation. Entresto (sacubitril/valsartan), β-blockers, and loop diuretics were prescribed to all patients.

Baseline characteristics
Among the 11 patients, six (54.5%) were male. All patients had cardiomyopathy (6 non-ischemic and 5 ischemic), and 3 patients had paroxysmal atrial brillation. Hypertension was present in 4 patients.
Frequent ventricular premature contraction (VPC) (> 1,000 per 24 hours[21]) were found in 3 patients. The mean age was 69.1 ± 6.4 years, and the baseline characteristics of the patients were provided in Table 1.
At baseline, the two groups were matched for age, gender, hypertension, diabetes mellitus, ICM, atrial brillation as illustrated in Table I (all P >0.05).

Procedural Outcomes
Biventricular ICDs (CRT-D) were implanted in 9 patients ( Figure 3A, 3B), and CRT pacemaker was implanted in the remaining 2 patients (Table 2). One in every group, CRTP did not differ between two groups (P > 0.05). The operation duration was 135 ± 26 min. The duration of X-ray uoroscopy was 29.2 ± 8.8 min. In group 1, LBBAP was successfully achieved in 5 patients. Another one failed and transferred into group 2. So the acute success rate was 83.3%.
In group 2, CS lead was successfully implanted in all 6 patients. The acute success rate was 100%. Compared with group 2, the operation duration and the duration of X-ray uoroscopy in group 1 all increased, but the later was not signi cantly (P > 0.05)( Table 2).
Both the LBBAP and LV capture thresholds remained stable during procedure (1.3 ± 0.6 V at 0.4 ms vs. 1.6 ± 0.7 V at 0.4 ms). Bipolar LBBAP resulted in partial but signi cant narrowing of the QRSd (BBB correction) in 5 patients.
Both groups did not show difference in CS pacing lead, RV de brillator lead parameters, such as R-wave amplitude, threshold, and impedance and so on ( Table 2).

ECG characteristics and pacing parameters
Individual electrocardiographic responses to RV, LV, and LBBAP at the time of implantation were shown in Table 2. Among the 11 patients, the baseline QRSd was 168.1 ± 18.9 ms (Figure 2a). At baseline, the two groups were matched for QRSd (158.0 ±13.0, vs. 176.7 ±19.7, P >0.05) as illustrated in Table 3.

Follow-up
The mean follow-up time was 300 ± 185 days. At baseline, the two groups were matched for follow-up time (296±201, 305±190, P >0.05). Among all 11 patients, CS lead parameters were stable during followup. In group 1, the LBBAP capture threshold, R-wave amplitude, and lead impedance were 0.74 ± 0.25 V, 13.36 ± 5.23 mV, and 533.73 ± 32.31 Ω during the 3-month follow-up (all P > 0.05, respectively, between the time of device implantation and the follow-up visit). In group 2, the RV lead parameters were also stable during follow-up. No patients showed signs of dislodgement, loss of capture, infections, embolism, or stroke associated with the implantation. The ventricular pacing rate was 95%. There were 8 VT/VF episodes treated with antitachycardia pacing and/or shock that had an electrogram available for adjudication (3 in group 1, 5 in group 2). However, the rate of VT/VF therapy was not statistically different (P =0 .175) between in 2 groups.
Transthoracic echocardiogram (Figure 3) evaluation data at baseline and at the 1-month and 3-month follow-ups were available in all 11 patients receiving successful CRT. As shown in As compared to the base line, patients in group 1 showed signi cant improvement in LVEF and NT-proBNP levels, while patients in group 2 showed non-signi cant changes in these parameters (Table 3).

Major ndings
The present study demonstrates the following merits. (1)  There is evidence that LV activation time is only minimally increased in RBBB but signi cantly increased in LBBB [26]. During unipolar LBBAP, as the right ventricle is predominantly activated via myocardial conduction, RV dyssynchrony may be present compared to HBP. However, it does not cause LV dyssynchrony since LV activation occurs via the His-Purkinje system. Therefore, in patients undergoing permanent LBBAP, synchronization of delayed RV activation and normal LV activation is feasible.

LOT-CRT advantage
However, in patients with intraventricular block or higher overall scar burden, success rates are somewhat limited depending on the site of block and the scar burden and distribution of the interventricular septum [9]. Intra-or interventricular dyssynchrony cannot be reduced through LBBAP. LBBAP achieved only partial reduction of the QRSd in those patients with a baseline surface ECG of atypical LBBB morphology [9]. LOT-CRT offers the advantage of using the LV lead in addition to LBBAP in a potential scenario in which conduction disease progresses.
In patients with LBBB and cardiomyopathy, LOT-CRT resulted in signi cant electrical resynchronization. In group 1 of our study, 60% of whose subjects had severe ischaemic cardiomyopathy, LOT-CRT resulted in high clinical and echocardiographic response rates. Our results indicated that patients with LBBB and a higher overall scar burden might be the desired candidates for LOT-CRT.

Limitations
First, LOT-CRT is time consuming. The duration of the operation was 152 ± 31 min, and the duration of Xray uoroscopy was 26.2 ± 5.9 min; both were longer than stated in a previous report (117 ± 48 and 16.4 ± 12.3 min)[8] and control group. Second, this study included only a small sample at a single centre. Third, this study had a short follow-up interval, although we expect favourable long-term clinical bene ts.
Furthermore, this study enrolled only 5 ischaemic patients. Although this study does not provide su cient data to support a general conclusion, we observed signi cant echocardiographic and clinical improvement in these HF patients treated with LOT-CRT.

Conclusions
The study demonstrates that LOT-CRT is clinically feasible in patients with systolic HF and LBBB. LOT-CRT was associated with signi cant reduction of QRS duration and improvement in LV function, especially in patients with ICM.

Consent for publication
Not applicable.

Availability of data and materials
Data are available from the corresponding author upon reasonable request due to privacy or other restrictions.
Funding none Authors' contributions XFF was the study advisor. XFF and RZ designed the study. RZ collected data and performed TTE. BL analysed the data. XFF, RZ, BL, YQH, and QFL performed the procedures. RZ was the main investigator and provided the rst draft. YGL critically revised the manuscript. XFF provided the nal draft. All authors read and approved the nal manuscript.