Left ventricular‐only fusion pacing versus cardiac resynchronization therapy in heart failure patients: A randomized controlled trial

Abstract Background It is unclear whether clinical benefits of cardiac resynchronization can be achieved by pacing only the left ventricle. Hypothesis We aimed to compare the effect of a novel adaptive left ventricular‐only fusion pacing (LVP) on ventricular function with conventional biventricular pacing (BVP) in cardiac resynchronization therapy (CRT) indicated patients. Methods This prospective, randomized, multicenter study enrolled CRT‐indicated patients with PR interval ≤ 200 ms who were randomized in the adaptive LVP group (using the AdaptivCRT™ algorithm with intentional non‐capture right ventricular pacing) or the echocardiography‐optimized BVP group. Cardiac function and echocardiography were evaluated at baseline and follow‐ups. CRT super response was defined as two‐fold or more increase of left ventricular ejection fraction (LVEF) or final LVEF >45%, and LV end‐systolic volume (LVESV) decrease >15%, and New York Heart Association (NYHA) class improved by at least one level. Results Sixty‐three patients were enrolled in the study (LVP = 34 vs. BVP = 29). At 6‐month follow‐up, significant improvements in LVEF, LVESV, and NYHA class were observed in both groups. The CRT super response rate was significantly higher in patients with high‐percentage adaptive LV‐only pacing in LVP group (68.4%) than in BVP group (36.4%, p = .04). Conclusions Adaptive LV‐only pacing was comparable to BVP in improving cardiac function and clinical condition in CRT‐indicated patients. This finding raises the possibility that an adaptive LVP algorithm with appropriate right ventricular sensing to fuse with intrinsic right ventricular activation in a two‐lead (right atrium and left ventricle) device may provide clinical benefit in a subset of CRT patients with intact atrioventricular conduction.

Recently, LV-only pacing is proposed as an alternative approach to achieve resynchronization by the fusion of LV pacing with intrinsic RV conduction. 3 Although acute and short-term investigations suggested comparable benefits of LV-only pacing and conventional biventricular pacing (BVP), [4][5][6] varying AV conduction may still lead to electrical dyssynchrony during various daily activities or due to the changes in disease state over a longer period of time. Therefore, an adaptive algorithm is developed to optimize the fusion by continuously adjusting LV pacing timing to leverage intrinsic RV conduction and achieve dynamic and more physiological pacing. The purpose of this study is to assess mid-term cardiac function and clinical outcomes during adaptive LV-only fusion pacing (LVP) in comparison with conventional BVP in a selected group of CRT-indicated patients with intact AV conduction.

| Study population
Patients were recruited in five centers. Inclusion criteria were (1)

| Study design
This study is a prospective, multicenter, randomized, controlled clinical trial. Consented patients who fulfilled inclusion and exclusion criteria were enrolled and randomly assigned to LVP or BVP group. Randomization was performed among all patients. Centralized random sequence was generated from natural number with odd or even number standing for LVP or BVP group, then site staff would assign each patient to a corresponding group and intervention per random sequence. Ethics committee approvals were obtained in all study sites.
Written informed consent was obtained from each patient. The study was registered at http://www.clinicaltrials.gov under the identifier NCT03071978.
All enrolled patients received CRT devices featured with the AdaptivCRT™ algorithm (Medtronic Inc., Minneapolis, MN) according to clinical demands. The AdaptivCRT algorithm 7-10 is a novel pacing algorithm for CRT by dynamically optimizing the AV and interventricular (VV) delays minute-by-minute based on the electrical conduction intervals. Furthermore, for patients with normal AV conduction, the AdaptivCRT algorithm recruits the intrinsic conduction and avoids providing RV pacing. For patients in LVP group, AdaptivCRT pacing mode was enabled after implantation to provide maximum fusion of LV pacing with intrinsic RV activation, while pacing through RV lead was functionally turned off (set as the minimal pacing parameters to make sure of non-capture RV pacing) to achieve the LV-only pacing. For patients in BVP group, AdaptivCRT was disabled after implantation, then AV and VV delays were optimized before discharge using echocardiographic evaluation with the method described by Gorcsan et al. 11 The pacing mode settings of enrolled patients will be determined by physician according to the clinical status after completing the 6-month follow-up and exiting this study. Echocardiographic measurements were performed with a commercially available system (Vingmed Vivid 7; GE Vingmed, Milwaukee, WI). Echocardiographic data were recorded for at least three consecutive cardiac cycles. LVEF, LVESV, and LVEDV were measured by Simpson's biplane method from the apical four-chamber view.

| Clinical and cardiac function assessments
Parameters were analyzed with a consistent protocol in a core laboratory. CRT response was defined as an absolute increase of LVEF >10% or a relative decrease of LVESV >15% or NYHA class improved by at least one level at 6-month follow-up compared to baseline value.
In addition, CRT super response 12 was defined as the composite score of a two-fold or more increase of LVEF or a final LVEF >45%, and LVESV decrease >15%, and NYHA class improved by at least one level at 6 months.

| Statistical analysis
Continuous variables were presented as means ± standard deviation, and categorical variables as numbers and proportions. Continuous variables were compared between baseline and 3-/6-month follow-up using paired Student's t-test or Wilcoxon signed-rank test. For comparisons of continuous variables between the LVP and BVP groups, independent Student's t-test or Mann-Whitney U test was applied.
Categorical variables were compared using the chi-square test or Fisher's exact test. Statistical significance was defined as a p < .05. All analyses were performed with SPSS software (version 22; SPSS, Chicago, IL).

| Patient characteristics
Sixty-three consecutive patients were prospectively enrolled from April 2017 to June 2018 and randomly assigned to LVP group (n = 34) or echocardiography-optimized BVP group (n = 29). Study follow-ups were completed in Jan 2019. The numbers of loss of follow-up at 6 months were three in LVP group and two in BVP group ( Figure 1), which included one mortality in each group. The cause of the two deaths was lung cancer. Characteristics of the study population at baseline are summarized in Table 1

| LVP subgroup analysis
In this study, the AdaptivCRT algorithm was enabled in LVP group to avoid RV pacing. Of 31 patients in LVP group, 25 had a high percentage of adaptive LV pacing that well fused with intrinsic RV activation (high-aLVP% subgroup; average LV-pacing percentage = 88.7%, average total ventricular pacing percentage = 96.1%) within 6 months. On the contrary, the remaining six patients had an average BVP percentage = 88.5% (average total ventricular pacing percentage = 95.5%) because AdaptivCRT algorithm could automatically switch the pacing mode to BVP when intrinsic RV conduction was failed to sense within default periodical check. Given that RV pacing output was set to the minimum for nonfunctional pacing, these six patients received LV-only pacing without fusion with intrinsic RV activation (low-aLVP% subgroup).
Subgroup analysis revealed significant improvement of LVEF and LVESV at 6 months in both subgroups, compared to baseline, respectively. However, significant improvements in QRSd, NYHA class, and 6MWD were observed in high-aLVP% at 6 months, but not in low-aLVP% subgroup. Furthermore, comparing to the BVP group, the high-aLVP% subgroup had significantly greater improvements in LVEF, NYHA (both p < .01) and QRSd (p < .05) at 6 months and a trend of better decrease in LVESV (p = .15; Figure 3), while no comparable improvements were observed in the low-aLVP% subgroup compared to the BVP group (LVEF increase 11.3% ± 6.9% vs. 10.8% With applying the definition of CRT super response, a trend of higher CRT super response rate was found in LVP group at 6 months T A B L E 1 Demographic data and baseline characteristics of the study population Note: Values are expressed as n or as mean ± standard deviation. p-value: LVP group versus BVP group. Abbreviations: 6MWD, 6-minute walking distance; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; CRT-P(%), percentage of patients who were implanted with CRT-pacemaker; LVEDD, left ventricular end-diastolic dimension; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESD, left ventricular end-systolic dimension; LVESV, left ventricular end-systolic volume; NT-pro BNP, N-terminal pro-brain natriuretic peptide; NYHA class, New York Heart Association class. a p value by Mann-Whitney U test of baseline PR.
compared with BVP group though statistical significance was not identified (p = .08, Figure 3(E)). However, the high-aLVP% subgroup resulted in a significantly larger CRT super response rate than the BVP group (p = .04, Figure 3(F)).

| DISCUSSION
The results of the present study demonstrated that LVP provides sig-

| Clinical perspectives
This study is the first prospective randomized multicenter clinical study to assess the feasibility of LV-only pacing fused with intrinsic RV activation in CRT-indicated patients.

| CONCLUSIONS
The present study demonstrated comparable clinical outcomes between adaptive LVP group and conventional BVP group. Moreover, our study found that high-percentage adaptive LVP was significantly associated with better clinical outcomes and a higher CRT super response rate than BVP. This finding raises the possibility that an adaptive LVP algorithm with appropriate RV sensing to fuse with intrinsic RV activation in a two-lead (right atrium and LV) device may provide clinical benefit in a subset of CRT patients with intact AV conduction.

DATA AVAILABILITY STATEMENT
On reasonable request, the datasets used and/or analyzed of the study will be available from the corresponding author.