Advertisement

On-treatment comparison between corrective His bundle pacing and biventricular pacing for cardiac resynchronization: A secondary analysis of the His-SYNC Pilot Trial

      Background

      The His-SYNC pilot trial was the first randomized comparison between His bundle pacing in lieu of a left ventricular lead for cardiac resynchronization therapy (His-CRT) and biventricular pacing (BiV-CRT), but was limited by high rates of crossover.

      Objective

      To evaluate the results of the His-SYNC pilot trial utilizing treatment-received (TR) and per-protocol (PP) analyses.

      Methods

      The His-SYNC pilot was a multicenter, prospective, single-blinded, randomized, controlled trial comparing His-CRT vs BiV-CRT in patients meeting standard indications for CRT (eg, NYHA II–IV patients with QRS >120 ms). Crossovers were required based on prespecified criteria. The primary endpoints analyzed included improvement in QRS duration, left ventricular ejection fraction (LVEF), and freedom from cardiovascular (CV) hospitalization and mortality.

      Results

      Among 41 patients enrolled (aged 64 ± 13 years, 38% female, LVEF 28%, QRS 168 ± 18 ms), 21 were randomized to His-CRT and 20 to BiV-CRT. Crossover occurred in 48% of His-CRT and 26% of BiV-CRT. The most common reason for crossover from His-CRT was inability to correct QRS owing to nonspecific intraventricular conduction delay (n = 5). Patients treated with His-CRT demonstrated greater QRS narrowing compared to BiV (125 ± 22 ms vs 164 ± 25 ms [TR], P < .001;124 ± 19 ms vs 162 ± 24 ms [PP], P < .001). A trend toward higher echocardiographic response was also observed (80 vs 57% [TR], P = .14; 91% vs 54% [PP], P = .078). No significant differences in CV hospitalization or mortality were observed.

      Conclusions

      Patients receiving His-CRT on-treatment demonstrated superior electrical resynchronization and a trend toward higher echocardiographic response than BiV-CRT. Larger prospective studies may be justifiable with refinements in patient selection and implantation techniques to minimize crossovers.

      Graphical abstract

      Keywords

      Cardiac resynchronization therapy (CRT) has an established role in the treatment of patients with heart failure and electromechanical dyssynchrony with wide QRS duration.
      • Bristow M.R.
      • Saxon L.A.
      • Boehmer J.
      • et al.
      Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure.
      • Cleland J.G.
      • Daubert J.C.
      • Erdmann E.
      • et al.
      The effect of cardiac resynchronization on morbidity and mortality in heart failure.
      • Moss A.J.
      • Hall W.J.
      • Cannom D.S.
      • et al.
      Cardiac-resynchronization therapy for the prevention of heart-failure events.
      As an adjunct to guideline-directed medical therapy, CRT has been consistently shown to improve quality of life, NYHA status, and left ventricular (LV) remodeling with improvements in LV ejection fraction (LVEF) and reduction in LV dimensions and volumes.
      • Cazeau S.
      • Leclercq C.
      • Lavergne T.
      • et al.
      Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay.
      Mortality reduction was demonstrated in the CARE HF trial with CRT pacing alone.
      • Cleland J.G.
      • Daubert J.C.
      • Erdmann E.
      • et al.
      The effect of cardiac resynchronization on morbidity and mortality in heart failure.
      Despite these benefits, nonresponse to CRT remains high, estimated between 30% and 40%.
      • Daubert C.
      • Behar N.
      • Martins R.P.
      • Mabo P.
      • Leclercq C.
      Avoiding non-responders to cardiac resynchronization therapy: a practical guide.
      European Heart Rhythm Association (EHRA)European Society of Cardiology (ESC)
      Heart Rhythm Society, et al. 2012 EHRA/HRS expert consensus statement on cardiac resynchronization therapy in heart failure: implant and follow-up recommendations and management.
      A possible limitation to conventional CRT is that electrical synchronization via biventricular pacing (BiV) is achieved through nonphysiological means, via fusion of an epicardial LV wavefront with an endocardial wavefront from the right ventricular apex.
      His bundle pacing (HBP) has been to shown to be a viable bailout option for CRT and more recently has been evaluated for feasibility as a first-line strategy.
      • Barba-Pichardo R.
      • Manovel Sanchez A.
      • Fernandez-Gomez J.M.
      • et al.
      Ventricular resynchronization therapy by direct His-bundle pacing using an internal cardioverter defibrillator.
      • Lustgarten D.L.
      • Calame S.
      • Crespo E.M.
      • et al.
      Electrical resynchronization induced by direct His-bundle pacing.
      • Lustgarten D.L.
      • Crespo E.M.
      • Arkhipova-Jenkins I.
      • et al.
      His-bundle pacing versus biventricular pacing in cardiac resynchronization therapy patients: A crossover design comparison.
      • Ajijola O.A.
      • Upadhyay G.A.
      • Macias C.
      • Shivkumar K.
      • Tung R.
      Permanent His-bundle pacing for cardiac resynchronization therapy: Initial feasibility study in lieu of left ventricular lead.
      • Sharma P.S.
      • Dandamudi G.
      • Herweg B.
      • et al.
      Permanent His-bundle pacing as an alternative to biventricular pacing for cardiac resynchronization therapy: A multicenter experience.
      • Huang W.
      • Su L.
      • Wu S.
      • et al.
      Long-term outcomes of His bundle pacing in patients with heart failure with left bundle branch block.
      Capture of the native conduction system can achieve complete restoration of normal physiologic His-Purkinje conduction, which may more favorably promote remodeling compared to BiV. Recent data suggest that the underlying pathophysiology of left bundle branch block (LBBB) patterns is attributable to focal disease located proximally in the left conduction system, which provides a mechanistic explanation for QRS correction when proximal pacing circumvents a site of distal conduction block with a sufficient pacing stimulus.
      • Upadhyay G.A.
      • Cherian T.
      • Shatz D.Y.
      • et al.
      Intracardiac delineation of septal conduction in left bundle branch block patterns: mechanistic evidence of left intra-hisian block circumvented by His pacing.
      Although there has been growing enthusiasm for HBP as a means to electrically correct wide QRS complexes and electrical cardiac resynchronization, only a single study reported outcomes comparing BiV-CRT vs HBP in lieu of an LV lead for CRT (His-CRT) in a crossover study design.
      • Lustgarten D.L.
      • Crespo E.M.
      • Arkhipova-Jenkins I.
      • et al.
      His-bundle pacing versus biventricular pacing in cardiac resynchronization therapy patients: A crossover design comparison.
      The His Bundle Pacing versus Coronary Sinus Pacing for Cardiac Resynchronization Therapy (His-SYNC) pilot trial was an investigator-initiated, prospective, randomized, controlled clinical trial that aimed to assess the feasibility and efficacy of His-CRT as a first-line strategy for CRT compared to BiV pacing for cardiac resynchronization (BiV-CRT) with regard to both electrocardiographic and echocardiographic response. By intention-to-treat analysis, His-CRT did not demonstrate improvements in electrocardiographic or echocardiographic parameters as compared to BiV-CRT.
      • Upadhyay G.A.
      • Vijayaraman P.
      • Nayak H.M.
      • et al.
      His Corrective Pacing or Biventricular Pacing For Cardiac Resynchronization In Heart Failure: A Randomized Pilot Trial (HIS-SYNC).
      Interpretation of these results, however, was confounded by high rates of crossover, which were mandated by protocol when CRT could not be achieved by the allocated treatment. We present secondary on-treatment analysis of His-SYNC to examine the direct effect of treatment completed and to describe reasons for crossover.

      Methods

      The study was conducted at 7 centers and the University of Chicago (UChicago) served as the Study Coordinating Site (NCT0270045). Approval by the local institutional review board was obtained at each participating center prior to enrollment, and all patients provided written informed consent. Patients were blinded to their treatment allocation (His-CRT or BiV-CRT) in this 2-parallel-arm study. Enrollment began in May 2016 and ended in June 2018. All data were sent to the core laboratory (UChicago) for analysis. An interim analysis was performed midway (January 2018) through enrollment to assess for safety, and no significant differences were noted.
      Eligible patients with heart failure (HF) greater than 18 years of age meeting ACCF/AHA/HRS class 1 or class II guideline indications for CRT were considered for inclusion.
      • Epstein A.E.
      • DiMarco J.P.
      • Ellenbogen K.A.
      • et al.
      2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society.
      Exclusion criteria included existing CRT device, pregnancy, or inability of the patient to provide consent for himself or herself owing to either medical or psychiatric comorbidity. Subjects were not compensated for participation.

      Study procedures

      The implant procedure was performed as per standard technique for cardiac implantable electronic devices. Patients were centrally randomized to assignment to His-CRT or traditional coronary sinus (CS) lead for BiV-CRT. HBP was performed utilizing the Medtronic SelectSecure Model 3830 lead (Medtronic, Minneapolis, MN). Delivery of the lead was performed utilizing either the fixed-curve Model C315HIS catheter (Medtronic, Minneapolis, MN) or the deflectable Model C304 sheath (Medtronic, Minneapolis, MN), per the operator preference. His bundle mapping and lead fixation at the time of implant was performed as described previously.
      • Ajijola O.A.
      • Upadhyay G.A.
      • Macias C.
      • Shivkumar K.
      • Tung R.
      Permanent His-bundle pacing for cardiac resynchronization therapy: Initial feasibility study in lieu of left ventricular lead.
      Briefly, His bundle mapping was performed using standard fluoroscopic views to assess septal orientation in left anterior oblique and anterosuperior orientation in right anterior oblique. The sheath was positioned across the tricuspid annulus and the helix of the lead was introduced beyond the sheath to assess local electrograms guided by the atrial-ventricular ratio. The His potential was mapped with counterclockwise rotation of the sheath to direct inferoposteriorly toward the septum and clockwise to direct anterosuperiorly away from the septum. Pacing was performed with unipolar and bipolar configuration to assess for QRS narrowing at sites with stable His potential recording.
      Patients allocated to LV lead for BiV-CRT underwent CS cannulation and LV lead placement per routine implant procedure.
      • Prinzen F.W.
      • Vernooy K.
      • Auricchio A.
      Cardiac resynchronization therapy: state-of-the-art of current applications, guidelines, ongoing trials, and areas of controversy.
      LV lead, type, and vendor (eg, Abbott, Biotronik, Boston Scientific, Medtronic, or MicroPort) were left to the discretion of the implanting physician. Intraprocedurally, operators were encouraged to target a posterolateral or lateral branch of the CS in regions with long electrical delay as assessed by surface QRS-to-LV sense (QLV) timing. Placement of the LV lead into the anterior interventricular vein or middle cardiac vein was discouraged. In order to facilitate optimal lead placement, arterial access for levo-phase CS angiography or LV septal mapping to characterize site of bundle branch block was also permitted at the discretion of the implanting physician. Crossover was encouraged in patients randomized to BiV-CRT when an LV lead could not be placed owing to difficult cannulation of the CS or limited CS branch targets, or when phrenic nerve capture resulted in diaphragmatic stimulation during BiV. V-V interval timing was adjusted by the implanting physician to optimize QRS narrowing.
      Crossover was mandated in patients randomized to His-CRT if the paced QRS width did not narrow by at least 20% or to a QRS width of ≤130 ms (owing to data showing lack of benefit for BiV at QRS widths below this cut-off),
      • Ruschitzka F.
      • Abraham W.T.
      • Singh J.P.
      • et al.
      Cardiac-resynchronization therapy in heart failure with a narrow QRS complex.
      or if fixation of the HBP lead could not be performed with adequate stability or pacing output (≤5 V @ 1.0 ms). QRS correction with either nonselective or selective capture was accepted, as described in a recent working group recommendations statement.
      • 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.
      Corrected QRS width was measured from the onset of the intrinsicoid R wave noted in V1 or V2, as described based on data regarding His engagement from intracardiac left-sided recordings.
      • Upadhyay G.A.
      • Cherian T.
      • Shatz D.Y.
      • et al.
      Intracardiac delineation of septal conduction in left bundle branch block patterns: mechanistic evidence of left intra-hisian block circumvented by His pacing.
      An example of a patient meeting successful QRS correction is shown in Figure 1.
      Figure thumbnail gr1
      Figure 1Typical left bundle branch block pattern (QRS width 195 ms, top) with successful QRS correction (bottom) from nonselective His capture resulting in QRS narrowing to 125 ms, as measured utilizing the onset of the intrinsicoid R wave in lead V2.
      Patients were followed with an incision check at approximately 2 weeks and routine clinical evaluation at 1, 3, 6, and 12 months. Patients underwent device interrogation in person or remotely at each of the available time points. Reported device measurements included lead sensing (mV), pacing capture threshold (V), pulse width (ms), and impedance (ohms). In patients with multipoint pacing enabled (n = 2), measurements utilizing the lowest LV bipole stimulation output were used for comparison.

      Study endpoints

      The primary endpoints included measures of electrocardiographic (ie, change in QRS width) and echocardiographic (ie, change in LVEF) parameters at 6 months. Time to first cardiovascular (CV) hospitalization or all-cause mortality at 12 months was also examined. Patient-level electrocardiography (ECG) and echocardiographic files were sent to the UChicago core laboratory for adjudication and reassessment.
      QRS widths were measured using electronic calipers (Cardio Calipers, Iconico, Philadelphia, PA). Patients with baseline bundle branch block morphology were categorized according to the classification set forth in the AHA/ACCF/HRS guidelines.
      • Surawicz B.
      • Childers R.
      • Deal B.J.
      • et al.
      AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology.
      Patients with LBBB pattern that did not meet the Strauss criteria
      • Strauss D.G.
      • Selvester R.H.
      • Wagner G.S.
      Defining left bundle branch block in the era of cardiac resynchronization therapy.
      (ie, QRS width ≥130 ms in women, ≥140 in men; QS or rS in leads V1 and V2; and mid-QRS notching or slurring in 2 of leads V1, V2, V5, V6, I, and aVL) were categorized as intraventricular conduction delay (IVCD).
      Echocardiographic assessment was performed by 2 readers blinded to the type of device received. Baseline and follow-up (6-month) transthoracic echocardiography were subsequently analyzed offline using the Excelera software platform (Philips Healthcare, Andover, MA). Left ventricular end-diastolic volume (LVEDV) and left ventricular end-systolic volume (LVESV) were calculated using the disk summation measured from the 4-chamber and 2-chamber views with LVEF calculated using the standard formula (ie, LVEF = [LVEDV − LVESV]/LVEDV). Echocardiographic response was defined as an LVEF improvement by ≥5% from baseline.
      • Vijayaraman P.
      • Herweg B.
      • Ellenbogen K.A.
      • Gajek J.
      His-optimized cardiac resynchronization therapy to maximize electrical resynchronization.
      Secondary outcomes included incidence of CV hospitalization alone and incidence of significant ventricular arrhythmia (ie, ventricular tachycardia [VT] or ventricular fibrillation [VF] requiring device therapy) during 12 months of follow-up. Quality of life was also assessed utilizing the Kansas City Cardiomyopathy (KCCQ) questionnaire, which was administered at baseline and at 6 and 12 months, along with assessment of NYHA function class. Latest available KCCQ data at 6 or 12 months were used for analysis. Data regarding endpoints of interest were collected and stored prospectively a remote database (REDCap, Vanderbilt University, Nashville, TN).
      Procedure-related complications were reported, including those that were implant-related (ie, pneumothorax, perforation, pericardial effusion, implant site hematoma, implant site infection) and lead-related (ie, lead dislodgment, lead fracture, or inability to pace owing to high threshold or phrenic capture) during follow-up.

      Statistical analysis

      For baseline and clinical characteristics, continuous variables were expressed as means ± standard deviations or medians with interquartile ranges and compared with either independent t tests, Wilcoxon signed rank tests, or Mann-Whitney U tests, depending upon normality. The sample size was estimated based on the primary endpoint of echocardiographic response to test the hypothesis that an absolute 10% greater improvement in LVEF would be observed with His-CRT compared to BiV, with a significance level (α) = 0.05 and a power of 0.8. Categorical variables were expressed as relative counts and percentages and compared with χ2 tests of association or Fisher exact tests, as appropriate. Kaplan-Meier curves were generated to describe time to CV hospitalizations and mortality, and then tested using log-rank tests. All statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc, Cary, NC) and STATA MP version 15 (StataCorp, College Station, TX).

      Results

      Study population

      A total of 41 patients were enrolled; 21 were initially randomized to His-CRT and 20 to BiV-CRT, with 1 patient withdrawal prior to device implantation in the BiV arm. The average duration of follow-up was 12.2 months, with 1 patient lost to follow-up during the study period. Overall, the mean age was 64.6 ± 12.6 years, 38% were female, 63% were white, 65% demonstrated a history of coronary artery disease, 33% had a history of paroxysmal or persistent atrial fibrillation, and 48% had a history of chronic kidney disease. Patients were maintained on guideline-directed medical therapy with 98% on beta-blockade and 75% on either angiotensin-converting enzyme inhibitor, angiotensin II receptor blocker, or combination with neprilysin inhibitor.
      Among the 21 patients that were randomized to His-CRT, 11 patients received successful HBP implantation with correction of QRS. Among 20 randomized to BiV-CRT, 1 patient withdrew prior to implantation, and 14 patients received successful LV lead implantation. High rates of crossover were noted in both arms (10/21 [48%] patients assigned to His-CRT and 5/19 [26%] BiV-CRT) and was not significantly different between the 2 arms (P = .20). Analysis according to treatment received (TR) consisted of 16 His-CRT and 24 BiV-CRT patients, and per-protocol (PP) analysis included 11 His-CRT and 14 BiV-CRT patients (Figure 2). No significant differences were noted in His-CRT vs BiV-CRT based on baseline demographics when analyzed as treated (Table 1). The mean procedural time was 2.9 ± 1.3 hours for His-CRT and 2.4 ± 1.2 hours for BiV-CRT (P = .25). Patients that necessitated crossover had significantly longer procedural times than those that were implanted PP (3.3 ± 1.6 hours vs 2.3 ± 0.9 hours, P = .04).
      Figure thumbnail gr2
      Figure 2Flow chart of patients in study. BiV = biventricular pacing; CRT = cardiac resynchronization therapy; CS = coronary sinus; IVCD = intraventricular conduction delay.
      Table 1Baseline demographic, electrocardiographic, and echocardiographic characteristics of patients by treatment received
      CharacteristicHis-CRT (n = 16)BiV-CRT (n = 24)P value
      Sex7 (43.8%)8 (33.3%).51
      Age63.4 ± 13.365.5 ± 12.4.61
      Height (cm)167.3 ± 11.9171.2 ± 11.6.31
      Weight (kg)84.0 (74.9–88.5)83.5 (77.6–96.0).81
      BMI28.8 (26.4–33.2)29.7 (26.2–31.3).99
      Race
       White10 (62.5%)15 (62.5%)1
       African-American5 (31.2%)7 (29.2%)1
       Hispanic1 (6.2%)1 (4.2%)1
       Other0 (0.0%)1 (4.2%)1
      Medication use
       BB15 (93.8%)24 (100%).40
       ACE-I4 (25.0%)7 (29.2%)1
       ARB4 (25.0%)6 (25.0%)1
       ARB/neprilysin inhibitor4 (25.0%)5 (20.8%)1
       Amiodarone1 (6.3%)7 (29.2%).11
       Digoxin0 (0.0%)6 (25.0%).06
       Spironolactone7 (43.8%)7 (29.2%).34
      Comorbidities
       HTN11 (68.8%)19 (79.2%).48
       CAD10 (62.5%)16 (6.7%).80
       History of CABG2 (12.5%)5 (20.8%).68
       History of AF4 (25.0%)9 (37.5%).42
       COPD4 (25.0%)6 (25.0%)1
       DM28 (50.0%)11 (45.8%).80
       CKD8 (50.0%)11 (45.8%).80
       ESRD1 (6.7%)2 (8.3%)1
       NYHA class3.0 (2.25–3.0)2.75 (2.25–3.0).66
      Electrocardiographic characteristics
       PR183 (166–199)192 (162–244).32
       QRS174 ± 18165 ± 17.12
       QTc485 ± 41479 ± 44.65
      Echocardiographic characteristics
       LVEDV184 (163–241)215 (171–271).29
       LVESV130 (110–180)157 (116–195).34
       LVEF28.0 (23.0–34.0)27.7 (23.6–30.7).81
      ACE-I = angiotensin-converting enzyme inhibitor; AF = atrial fibrillation; ARB = angiotensin II receptor blocker; BB = beta-blocker; BMI = body mass index; CAD = coronary artery disease (>70% stenosis of any vessel); CABG = coronary-artery bypass graft; CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; DM2 = diabetes mellitus type 2; ESRD = end-stage renal disease; LVEDV = left ventricular end-diastolic volume; LVESV = left ventricular end-systolic volume; LVEF = left ventricular ejection fraction; NYHA = New York Heart Association; PR = PR interval; QRS = QRS duration; QTc = corrected QT interval.

      Reasons for crossovers

      Five patients that crossed over to BiV-CRT from His-CRT had IVCD (not meeting Strauss criteria) could not be corrected with narrowing by at least 20% and to a QRS width ≤130 ms (Figure 3). Three patients demonstrated incomplete QRS narrowing with persistent width >130 ms. In 2 patients, the His bundle could not be mapped. Intraprocedural examples of patients demonstrating QRS narrowing and no correction are shown (Supplemental Figure 1). Left septal recordings were available for a subset of patients (n = 14), and corrective HBP was most likely in patients with either left intrahisian or block of proximal left bundle branch. HBP failed to correct any patient with intact Purkinje activation noted on left septal mapping consistent with IVCD. One patient randomized to His-CRT crossed over to BiV-CRT, and the CS could not be cannulated; the patient then underwent surgical LV lead placement guided by epicardial LV electroanatomic mapping to a region of long delay.
      Figure thumbnail gr3
      Figure 3Examples of 3 patients with IVCD that necessitated crossed over from His bundle pacing to biventricular pacing due to inability to correct QRS. All patients had QRS >130 ms but did not meet Strauss criteria for LBBB. A: 66 yo male with NICM, EF 34%, QRS 157 ms that improved to 41% with biventricular pacing at 6 months. B: 52 yo female with NICM, EF 20%, QRS 177 ms that improved to 30% with biventricular pacing at 5 months. C: 77 yo male with ICM EF 28% QRS 161 with mild improvement to 31% at 8 months with biventricular pacing. EF = ejection fraction; ICM = ischemic cardiomyopathy; LBBB = left bundle branch block; NICM = nonischemic cardiomyopathy; yo = year-old.
      In those that crossed over from BiV-CRT to His-CRT, reasons were as follows: suboptimal venous branch targets were present (n = 2), technical inability to cannulate the CS or target branch (n = 2), and vascular occlusion necessitating venoplasty (n = 1). Figure 4 shows an example of a patient with atretic CS anatomy without favorable targets who underwent successful His-CRT after crossover.
      Figure thumbnail gr4
      Figure 4Example of a patient randomized to biventricular cardiac resynchronization (BiV-CRT) that crossed over to His bundle pacing for CRT (His-CRT). A: Unfavorable anatomy is demonstrated by coronary sinus venography with atretic coronary sinus and unfavorable lateral vein targets. B: Baseline typical LBBB pattern is seen with QRS width of 170ms. C: Postimplant chest radiograph demonstrating 3-lead system with ICD in RV septum, His bundle lead, and atrial lead. D: 12-lead ECG postimplant shows successful resynchronization with His bundle pacing resulting in a final QRS duration of 120 ms. This patient demonstrated an echocardiographic super-response to His CRT from 35% to 53% in 5 months. ECG = electrocardiogram; ICD = implantable cardioverter defibrillator; LBBB = left bundle branch block; RV= right ventricular.

      Primary endpoints by on-treatment analysis

      Overall, the average QRS width was 168 ± 18 ms (35 LBBB pattern, 2 right bundle branch block, and 3 who were chronically right ventricular [RV] paced). Among patients with LBBB pattern at baseline, 71% met the Strauss criteria for LBBB. Patients undergoing His-CRT according to TR demonstrated greater electrical resynchronization than patients assigned to BiV-CRT. QRS width narrowed from baseline in His-CRT from 174 ± 18 ms to 125 ± 22 ms (P < .001), vs 165 ± 17 ms to 164 ± 25 ms for BiV-CRT (P = .82). QRS duration was significantly shorter in those that received His-CRT compared to those that received BiV-CRT (125 ± 22 ms vs 164 ± 25 ms; P < .001).
      There were no significant differences in LV volumes or LVEF at baseline in patients based on TR. At a median follow-up of 6.2 months, the LVEF improved significantly in both arms relative to baseline. Patients receiving His-CRT demonstrated a median increase in LVEF from 28.0% (23.0%-34.3%) to 34.6% (30.8%-45.0%) (P < .001). Patients that received BiV-CRT demonstrated a median increase in LVEF from 27.7% (23.6%-30.7%) to 32.0% (30.9%-40.1%) (P < .001). Pre- and post-QRS widths and LVEF are shown by group in Figure 5. In analysis by TR, median change in LVEF was numerically higher for His-CRT compared to BiV-CRT, but this difference was not statistically significant (+7.2% [5.0%-16.9%] vs +5.9% [1.5%-11.3%], P = .17). A trend toward higher rates of echocardiographic response (80% vs 57%, P = .14) was similarly observed. Decline in LVESV% was similar between both groups with a mean -22% ± 24% for those receiving His-CRT and -19% ± 14% (P = .62). Electrocardiographic and echocardiographic changes over the study are summarized in Figure 6.
      Figure thumbnail gr5
      Figure 5Pre and post QRS widths by individual patient in patients receiving His bundle pacing for CRT (His-CRT) versus biventricular pacing (BiV-CRT) shown in the top panel. Change in left ventricular ejection fraction (LVEF) before and after pacing is shown in lower panel.
      Figure thumbnail gr6
      Figure 6Change in QRS duration before and after pacing with BiV-CRT versus His-CRT, rate of echocardiographic response, and median change in LVEF before and after therapy by treatment-received group. BiV-CRT = biventricular pacing-cardiac resynchronization therapy; His-CRT = His bundle pacing-cardiac resynchronization therapy.
      Overall event rates were low during the study period, with a total of 6 CV hospitalizations (3 due to HF hospitalization, 2 periprocedural, and 1 for atrial fibrillation requiring cardioversion). Two patients demonstrated VT/VF requiring device therapy in follow-up, 1 of which resulted in pulseless electrical activity and death (BiV-CRT crossed over to His-CRT). One other death occurred outside the hospital; in this case, device interrogation data were unavailable (His-CRT). There were no significant differences between patients receiving His-CRT vs BiV-CRT with respect to the composite of time to CV hospitalization or all-cause mortality.

      Primary endpoints by per-protocol analysis

      In analysis of patients receiving treatment PP allocation, His-CRT was superior to BiV-CRT for electrical resynchronization (124 ± 19 ms vs 162 ± 24 ms; P < .001). No significant difference in median LVEF improvement was observed between groups (median increase in LVEF +11.8% His-CRT vs +5.2% BiV-CRT, P = .11), although numerically higher response was observed with His-CRT compared to as-treated analysis (Supplemental Figures 2 and 3). A stronger trend toward echocardiographic response was found in patients receiving His-CRT in PP comparison (91% vs 54%, P = .078). Only 1 patient that completed His-CRT PP did not demonstrate an improvement in LVEF. No difference in reduction in LVESV% was seen, with -22% ± 25% improvement with His-CRT vs -19% ± 17% with BiV-CRT (P = .75).

      Secondary endpoints

      Median NYHA class at baseline was comparable between patients treated with His-CRT (median NYHA class 3 [2.25–3.0]) vs BiV-CRT (median NYHA class 2.75 [2.25–3.0]) (P = .66). Improvement by ≥1 functional class was similar between the 2 groups at 6 months (53% His-CRT vs 39% BiV-CRT, P = .41) and at 12 months (25% His-CRT vs 31% BiV-CRT, P = .89). No patients declined by ≥1 functional grade at 6 months in either group. At 12 months, 1 patient declined by a single functional grade in the His-CRT group and none in the BiV-CRT group (P = 1.0). Median total KCCQ score at baseline was 101 points (78–112 points). Rise in KCCQ was noted for both patients receiving His-CRT (median +16 points [+8–25 points]) and BiV-CRT (median +10 points [+2–16 points]) and was not significantly different between the 2 groups (P = .22). No differences in VT/VF were observed during follow-up, with 2 events occurring in those with His-CRT and none in patients receiving BiV-CRT (P = .16).

      Adverse events

      A total of 4 periprocedural complications occurred. One patient with severe peripheral vascular disease and prolonged procedure (patient who received BiV-CRT after crossover) sustained a transient ischemic attack with aphasia, which resolved in follow-up. Two patients developed atrial lead microdislodgment associated with pericardial effusion (2 BiV-CRT, 1 of whom had crossed over). One hematoma in a patient receiving His-CRT required evacuation, without additional sequelae. No infectious complications were observed.

      Device parameters

      Device parameters at baseline and follow-up are noted in Table 2 according to the TR. No significant differences were found in right atrial or RV lead measurements of sensing, threshold, or impedance. QLV was reported in 20 of 24 patients receiving BiV-CRT across both arms (mean 131 ± 29 ms; mean QLV ratio 0.80 ± 0.19). With respect to resynchronization pacing, His-CRT was associated with higher pacing output to achieve QRS correction (2.75 V vs 0.85 V, P = .002) and pacing pulse width (1 ms vs 0.5 ms, P < .001) at baseline and in follow-up. No significant change was found in His lead output or pulse width at baseline when comparing baseline to 6 (2 V at 1 ms, P = .63 vs baseline) or 12 months (2.5 V at 1 ms, P = .99 vs baseline). Two patients treated with His-CRT were observed to have higher correction thresholds with loss of QRS correction in follow-up. In both cases, the His lead was maintained and the configuration of the pacing vector was changed from His tip–to–His ring to His tip–to–RV coil configuration, which reduced in correction threshold (2.25 V @ 1.0 ms and 3.75 V @ 1.0 ms).
      • Su L.
      • Xu L.
      • Wu S.J.
      • Huang W.J.
      Pacing and sensing optimization of permanent His-bundle pacing in cardiac resynchronization therapy/implantable cardioverter defibrillators patients: value of integrated bipolar configuration.
      No dislodgments of His or LV leads were seen during the 12-month study period in either group.
      Table 2Device parameters in follow-up by on-treatment group (His-CRT or BiV)
      ParameterHis-CRT (n = 16)BiV (n = 24)P value
      Implant measurements
       RA lead sensing (mV)2.00 (1.50–3.80)1.75 (1.25–3.50).457
       RA lead capture threshold (V)0.75 (0.60–1.25)0.70 (0.50–1.00).134
       RA lead pulse width (ms)0.40 (0.40–0.50)0.50 (0.40–0.50).384
       RA lead impedance (ohms)475 (399–726)494 (399–600).641
       RV lead sensing (mV)10.25 (9.40–14.80)12.00 (7.30–15.10).910
       RV lead threshold (V)0.50 (0.50–0.75)0.50 (0.50–1.00).241
       RV lead pulse width (ms)0.40 (0.40–0.50)0.40 (0.40–0.50).746
       RV lead impedance (ohms)513 (475–608)492.50 (428–555).334
      His or LV threshold (V)
      Threshold for QRS correction is reported for His bundle pacing.
      2.75 (1.25–3.38)0.85 (0.73–1.31).002
      His or LV pulse width (ms)1.00 (1.00–1.00)0.50 (0.40–0.65)<.001
      His or LV impedance (ohms)433 (340–481)540 (497–680).001
      6-Month follow-up
       RA lead sensing (mV)3.55 (2.50–4.60)2.40 (1.95–3.55).111
       RA lead capture threshold (V)0.75 (0.50–0.80)0.66 (0.55–0.81).721
       RA lead pulse width (ms)0.40 (0.40–0.40)0.40 (0.40–0.40).684
       RA lead impedance (ohms)485 (456–513)456 (380–513).345
       RV lead sensing (mV)13.65 (9.13–16.88)12.50 (11.30–20.00).275
       RV lead threshold (V)0.50 (0.50–0.75)0.75 (0.50–1.00).184
       RV lead pulse width (ms)0.40 (0.40–0.40)0.40 (0.40–0.40).647
       RV lead impedance (ohms)456 (399–551)428 (380–513).361
      His or LV threshold (V)
      Threshold for QRS correction is reported for His bundle pacing.
      2.00 (1.00–3.25)0.94 (0.75–1.25).004
      His or LV pulse width (ms)1.00 (1.00–1.00)0.40 (0.40–0.50)<.001
      His or LV impedance (ohms)295 (284–390)615 (456–703)<.001
      BiV = biventricular pacing; CRT = cardiac resynchronization therapy; LV = left ventricular; RA = right atrial; RV = right ventricular.
      Significant differences are bolded.
      Threshold for QRS correction is reported for His bundle pacing.

      Discussion

      The major findings of this secondary analysis of His-SYNC are as follows: (1) His-CRT was superior to BiV-CRT for electrical resynchronization, as measured by QRS narrowing; (2) echocardiographic response was numerically but not statistically higher in patients receiving His-CRT vs BiV-CRT; and (3) inclusion of IVCD accounted for the majority of crossovers in patients allocated to His-CRT.
      The His-SYNC pilot trial represents an investigator-initiated collaboration to prospectively assess the feasibility of His-CRT in comparison to BiV-CRT. Prior work has demonstrated the utility of His-CRT as a bailout strategy for failed attempt at traditional CRT and a single prospective investigation utilized Y-adapted His and LV leads to compare outcomes in a crossover design.
      • Lustgarten D.L.
      • Crespo E.M.
      • Arkhipova-Jenkins I.
      • et al.
      His-bundle pacing versus biventricular pacing in cardiac resynchronization therapy patients: A crossover design comparison.
      To the best of our knowledge, the His-SYNC pilot trial is the first randomized trial to compare His pacing in lieu of an LV lead with BiV-CRT in clinical practice. Further, it represents the first study in the field that prospectively compares two modalities to achieve CRT, but intention-to-treat analysis was confounded by high rates of crossover. To examine the physiologic effects of His-CRT, we performed secondary analysis based on on-treatment principles. PP analysis provides an assessment of the efficacy of His-CRT in an optimal setting.
      Importantly, His-SYNC compared echocardiographic response between two methods to achieve CRT rather than medical therapy. In contrast to the intention-to-treat analysis, in which patients assigned to His-CRT demonstrated lower baseline EF at baseline, the present analyses showed well-matched groups, with no differences in LV volumes, LVEF, or baseline demographics by TR. COMPANION and CARE-HF were the initial landmark trials to demonstrate mortality benefit in patients undergoing CRT with BiV compared to medical therapy.
      • Bristow M.R.
      • Saxon L.A.
      • Boehmer J.
      • et al.
      Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure.
      • Cleland J.G.
      • Daubert J.C.
      • Erdmann E.
      • et al.
      The effect of cardiac resynchronization on morbidity and mortality in heart failure.
      Improvements in LVEF have been inconsistently correlated with the magnitude of QRS narrowing,
      • Iler M.A.
      • Hu T.
      • Ayyagari S.
      • et al.
      Prognostic value of electrocardiographic measurements before and after cardiac resynchronization device implantation in patients with heart failure due to ischemic or nonischemic cardiomyopathy.
      although when dichotomized, increases in paced QRS duration
      • Menet A.
      • Bardet-Bouchery H.
      • Guyomar Y.
      • et al.
      Prognostic importance of postoperative QRS widening in patients with heart failure receiving cardiac resynchronization therapy.
      and duration over 200 ms
      • Hsing J.M.
      • Selzman K.A.
      • Leclercq C.
      • et al.
      Paced left ventricular QRS width and ECG parameters predict outcomes after cardiac resynchronization therapy: PROSPECT-ECG substudy.
      may predict poor response to BiV-CRT. Although there was no significant improvement in QRS duration in patients randomized to BiV-CRT in this pilot study, LVEF improvements were observed and long mean QLV timing reflected optimized LV lead positions. Importantly, reductions in mortality and LV volume indices have been demonstrated in pivotal trials, with relatively modest EF improvements. In CARE-HF, differences in mean EF between patients that received CRT compared with medical therapy were 3.7% at 3 months and 6.9% at 18 months. Similarly, differences in mean LVEF in MADIT CRT at 1 year between patients treated with ICD only vs ICD with CRT were 8% (3% vs 11%).
      • Moss A.J.
      • Hall W.J.
      • Cannom D.S.
      • et al.
      Cardiac-resynchronization therapy for the prevention of heart-failure events.
      The echocardiographic response for patients receiving His-CRT (median +7.2 [TR] and median +11.8 [PP]) is consistent with positive response with His-CRT, but did not reach the estimated 10% difference between groups (Supplemental Figure 4). In the present study, differences in medians of approximately 7% were observed in the PP analysis in favor of His-CRT over BiV-CRT (P = .078). Indeed, when examining response by initial ECG pattern, all but 1 patient who received His-CRT with LBBB pattern fulfilling Strauss criteria at baseline demonstrated echocardiographic response. These findings, however, should only be interpreted as hypothesis-generating owing to the as-treated comparison. When combined with previous data, these observations justify the need for larger prospective studies designed to test noninferiority and superiority of this novel technique for cardiac resynchronization. The present work also helps to inform the power and sample size calculations for future studies.
      Prespecified criteria in the study protocol mandated crossover, and rates were particularly high in patients originally randomized to His-CRT. In this regard, this pilot study should be interpreted as a comparison between two strategies. The main reason for crossover from BiV-CRT to His-CRT, which was higher than prior studies (26%), were primarily due to unfavorable cardiac venous anatomy. One contributing factor was that LV leads were not placed in anterior interventricular or middle cardiac veins, as these locations have been associated with modest responses to CRT. In addition, operators were conscious of maximizing electrical delay of implanted leads, as QLV was >95 ms in 85% of patients receiving BiV (mean QLV 131 ± 29 ms; mean QLV/QRS duration ratio 0.80 ± 0.19).
      Analysis of the baseline ECGs in patients that required crossover to BiV from His-CRT revealed that 50% of patients exhibited nonspecific IVCD. The inclusion criteria for this pilot trial encompassed a broad population based on current guidelines for traditional BiV-CRT. Subsequent to the design of His-SYNC, we have recently demonstrated that IVCD patients with intact Purkinje activation based on intracardiac mapping exhibit 0% predictive value for corrective His bundle pacing.
      • Upadhyay G.A.
      • Cherian T.
      • Shatz D.Y.
      • et al.
      Intracardiac delineation of septal conduction in left bundle branch block patterns: mechanistic evidence of left intra-hisian block circumvented by His pacing.

      Limitations

      This pilot study with small sample size was underpowered to detect differences less than 10% between groups, and the possibility of a type II error cannot be excluded. The criteria for LBBB definitions have been shown to clearly impact the outcomes of CRT studies and electrophysiologic definitions are essential to stratify patients that are most likely to benefit from resychronization.
      • Caputo M.L.
      • van Stipdonk A.
      • Illner A.
      • et al.
      The definition of left bundle branch block influences the response to cardiac resynchronization therapy.
      Implantation tools have not evolved substantially since 2010 with introduction of the C315HIS sheath. Longer helices, deflectable sheaths with septal orientation, and variable curves to accommodate variable patient anatomy are likely to improve His correction rates and stability of thresholds.

      Future directions

      The present results from this pilot study strongly suggest that His-CRT is not suitable for an unselected CRT population. With crossover rates approaching 50% from His-CRT, further investigation with the same study protocol is not warranted. The findings of this pilot study inform the methodology of the planned HIS SYNC II trial, where refinement in patient selection to exclude those with IVCD may reduce crossovers substantially. Future studies that evaluate His-CRT should target patients with an LBBB pattern that reflects complete conduction block that can be circumvented by His bundle pacing,
      • Upadhyay G.A.
      • Cherian T.
      • Shatz D.Y.
      • et al.
      Intracardiac delineation of septal conduction in left bundle-branch block patterns.
      and identify surface ECG characteristics that can consistently identify these patients. These mechanistic insights were discovered during the same time period in which the present trial was conducted. Additionally, advances in sheath and lead technology are necessary to improve implantation success, as current delivery tools do not adequately address patient variability in heart orientation and chamber dimensions. In addition, new approaches are being developed to engage the conduction system, including a novel intraseptal fixation technique that captures the left bundle branch and may improve the ability to correct distal block in the left bundle with lower thresholds.
      • 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.
      This technique expands options available for physiologic pacing in patients in whom corrective His bundle pacing is not achievable, but will require further prospective evaluation and validation.

      Conclusion

      In this first randomized pilot trial of His-CRT vs BiV for CRT in clinical practice, His-CRT demonstrated superior electrical resynchronization than BiV-CRT in on-treatment analysis, with a trend toward greater echocardiographic improvement, which did not reach significance. These secondary analyses should be interpreted as hypothesis-generating, and larger prospective studies with refinements in patient selection and implantation techniques may be justifiable to test for differential clinical outcomes between CRT modalities.

      Acknowledgments

      We would like to thank Dr Pratik Patel, Rush University Medical Center, Chicago, Illinois, and Dr Susan Kim, Northwestern University, Chicago, Illinois for their collaboration and assistance with enrollment.

      Appendix. Supplementary data

      References

        • Bristow M.R.
        • Saxon L.A.
        • Boehmer J.
        • et al.
        Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure.
        N Engl J Med. 2004; 350: 2140-2150
        • Cleland J.G.
        • Daubert J.C.
        • Erdmann E.
        • et al.
        The effect of cardiac resynchronization on morbidity and mortality in heart failure.
        N Engl J Med. 2005; 352: 1539-1549
        • Moss A.J.
        • Hall W.J.
        • Cannom D.S.
        • et al.
        Cardiac-resynchronization therapy for the prevention of heart-failure events.
        N Engl J Med. 2009; 361: 1329-1338
        • Cazeau S.
        • Leclercq C.
        • Lavergne T.
        • et al.
        Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay.
        N Engl J Med. 2001; 344: 873-880
        • Daubert C.
        • Behar N.
        • Martins R.P.
        • Mabo P.
        • Leclercq C.
        Avoiding non-responders to cardiac resynchronization therapy: a practical guide.
        Eur Heart J. 2017; 38: 1463-1472
        • European Heart Rhythm Association (EHRA)
        • European Society of Cardiology (ESC)
        Heart Rhythm Society, et al. 2012 EHRA/HRS expert consensus statement on cardiac resynchronization therapy in heart failure: implant and follow-up recommendations and management.
        Europace. 2012; 14: 1236-1286
        • Barba-Pichardo R.
        • Manovel Sanchez A.
        • Fernandez-Gomez J.M.
        • et al.
        Ventricular resynchronization therapy by direct His-bundle pacing using an internal cardioverter defibrillator.
        Europace. 2013; 15: 83-88
        • Lustgarten D.L.
        • Calame S.
        • Crespo E.M.
        • et al.
        Electrical resynchronization induced by direct His-bundle pacing.
        Heart Rhythm. 2010; 7: 15-21
        • Lustgarten D.L.
        • Crespo E.M.
        • Arkhipova-Jenkins I.
        • et al.
        His-bundle pacing versus biventricular pacing in cardiac resynchronization therapy patients: A crossover design comparison.
        Heart Rhythm. 2015; 12: 1548-1557
        • Ajijola O.A.
        • Upadhyay G.A.
        • Macias C.
        • Shivkumar K.
        • Tung R.
        Permanent His-bundle pacing for cardiac resynchronization therapy: Initial feasibility study in lieu of left ventricular lead.
        Heart Rhythm. 2017; 14: 1353-1361
        • Sharma P.S.
        • Dandamudi G.
        • Herweg B.
        • et al.
        Permanent His-bundle pacing as an alternative to biventricular pacing for cardiac resynchronization therapy: A multicenter experience.
        Heart Rhythm. 2018; 15: 413-420
        • Huang W.
        • Su L.
        • Wu S.
        • et al.
        Long-term outcomes of His bundle pacing in patients with heart failure with left bundle branch block.
        Heart. 2019; 105: 137-143
        • Upadhyay G.A.
        • Cherian T.
        • Shatz D.Y.
        • et al.
        Intracardiac delineation of septal conduction in left bundle branch block patterns: mechanistic evidence of left intra-hisian block circumvented by His pacing.
        Circulation. 2019; 139: 1876-1888
        • Upadhyay G.A.
        • Vijayaraman P.
        • Nayak H.M.
        • et al.
        His Corrective Pacing or Biventricular Pacing For Cardiac Resynchronization In Heart Failure: A Randomized Pilot Trial (HIS-SYNC).
        J Am Coll Cardiol. 2019; 74: 157-159
        • Epstein A.E.
        • DiMarco J.P.
        • Ellenbogen K.A.
        • et al.
        2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society.
        J Am Coll Cardiol. 2013; 61: e6-e75
        • Prinzen F.W.
        • Vernooy K.
        • Auricchio A.
        Cardiac resynchronization therapy: state-of-the-art of current applications, guidelines, ongoing trials, and areas of controversy.
        Circulation. 2013; 128: 2407-2418
        • Ruschitzka F.
        • Abraham W.T.
        • Singh J.P.
        • et al.
        Cardiac-resynchronization therapy in heart failure with a narrow QRS complex.
        N Engl J Med. 2013; 369: 1395-1405
        • 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
        • Surawicz B.
        • Childers R.
        • Deal B.J.
        • et al.
        AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology.
        Circulation. 2009; 119: e235-e240
        • Strauss D.G.
        • Selvester R.H.
        • Wagner G.S.
        Defining left bundle branch block in the era of cardiac resynchronization therapy.
        Am J Cardiol. 2011; 107: 927-934
        • Vijayaraman P.
        • Herweg B.
        • Ellenbogen K.A.
        • Gajek J.
        His-optimized cardiac resynchronization therapy to maximize electrical resynchronization.
        Circ Arrhythm Electrophysiol. 2019; 12: e006934
        • Su L.
        • Xu L.
        • Wu S.J.
        • Huang W.J.
        Pacing and sensing optimization of permanent His-bundle pacing in cardiac resynchronization therapy/implantable cardioverter defibrillators patients: value of integrated bipolar configuration.
        Europace. 2016; 18: 1399-1405
        • Iler M.A.
        • Hu T.
        • Ayyagari S.
        • et al.
        Prognostic value of electrocardiographic measurements before and after cardiac resynchronization device implantation in patients with heart failure due to ischemic or nonischemic cardiomyopathy.
        Am J Cardiol. 2008; 101: 359-363
        • Menet A.
        • Bardet-Bouchery H.
        • Guyomar Y.
        • et al.
        Prognostic importance of postoperative QRS widening in patients with heart failure receiving cardiac resynchronization therapy.
        Heart Rhythm. 2016; 13: 1636-1643
        • Hsing J.M.
        • Selzman K.A.
        • Leclercq C.
        • et al.
        Paced left ventricular QRS width and ECG parameters predict outcomes after cardiac resynchronization therapy: PROSPECT-ECG substudy.
        Circ Arrhythm Electrophysiol. 2011; 4: 851-857
        • Caputo M.L.
        • van Stipdonk A.
        • Illner A.
        • et al.
        The definition of left bundle branch block influences the response to cardiac resynchronization therapy.
        Int J Cardiol. 2018; 269: 165-169
        • Upadhyay G.A.
        • Cherian T.
        • Shatz D.Y.
        • et al.
        Intracardiac delineation of septal conduction in left bundle-branch block patterns.
        Circulation. 2019; 139: 1876-1888
        • 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: 1736.e1731-1736.e1733