Incidence of pacing‐induced cardiomyopathy in pacemaker‐dependent patients is lower with leadless pacemakers compared to transvenous pacemakers

Abstract Introduction Frequent right AQ4ventricular pacing (≥40%) with a transvenous pacemaker (TVP) is associated with the risk of pacing‐induced cardiomyopathy (PICM). Leadless pacemakers (LPs) have distinct physical and mechanical differences from TVP. The risk of PICM with LP is not known. To identify incidence, predictors, and long‐term outcomes of PICM in LP and TVP patients. Methods The study comprised all pacemaker‐dependent patients with LP or TVP who had left ventricular ejection fraction (LVEF) of ≥50 from 2014 to 2019. The incidence of PICM (≥10% LVEF drop) was assessed with an echocardiogram. Predictors for PICM were identified using multivariate analysis. Long‐term outcomes after cardiac resynchronization (CRT) were assessed in both groups. Results A total of 131 patients with TVP and 67 with LP comprised the study. All patients in the TVP group and the majority in the LP group underwent atrioventricular node ablation. The mean follow‐up duration in TVP and LP groups was 592 ± 549 and 817 ± 600 days, respectively. A total of 18 (13.7%) patients in TVP and 2 (3%) in LP developed PICM after a median duration of 254 (interquartile range: 470) days. The incidence of PICM was significantly higher with TVP compared with LP (p = .02). TVP as pacing modality was a positive (odds ratio [OR]: 1.07) while age was negative (OR: 0.94) predictor for PICM on multivariable analysis. Both patients in LP and all except two in the TVP group responded to CRT. Conclusion Incidence of PICM is significantly lower with LP compared with TVP in pacemaker‐dependent patients. Age and TVP as pacing modality were predictors for PICM.


| INTRODUCTION
Leadless pacemakers (LPs) offer an excellent alternative to transvenous pacemakers (TVPs) in patients needing single ventricular pacing and avoid the complications inherent to the transvenous devices including pocket infections, lead failures, chronic venous occlusions, tricuspid valve regurgitation, pocket hematomas, and pneumothoraces. 1,2 The Micra transcatheter catheter pacing system (Micra-TPS; Medtronic Inc.) is the only commercially available LP.
Before the recent approval of atrioventricular (AV) synchronous Micra (Micra AV), only 15% of the patients were eligible for LP.
Based on the existing LP use trend, most of the patients implanted with LP are either those who need infrequent ventricular pacing or those who have a normal ventricular function and undergo AV node ablation for atrial fibrillation with a rapid ventricular response.
Patients needing frequent right ventricle (RV) pacing have a 10%-15% incidence of pacing-induced cardiomyopathy (PICM). 3,4 Certain clinical predictors for PICM have been identified and include wide QRS duration at baseline, RV pacing burden, and reduced ventricular function before pacemaker implantation. Some studies have suggested a role of tricuspid regurgitation (TR) and apical location of pacing lead in the RV as contributory to the development of PICM; however, the data are discordant. 5,6 Incidence of PICM has not been reported for LP. There are several physical and mechanical differences between TVP and LP. LP is preferentially implanted in the septum and may be immune to lead-related development of TR. Implications of these factors for the development of PICM in patients with LP are not known.
This study aimed to compare the incidence and predictors of PICM in patients with LP and TVP.

| Patient population
The study comprised all patients with a normal left ventricular ejection fraction (LVEF ≥ 50%) at baseline who underwent LP or TVP implantation from February 2014 to June 2019. The data were obtained by a retrospective review of prospectively maintained databases for implantable devices at Ohio State University. The Institutional Review Board of the Ohio State University approved the study. Patients with a history of cardiac resynchronization therapy (CRT) and recovered LVEF who underwent extraction followed by single ventricle pacing were excluded from the study. In addition, patients who were not 100% pacemaker dependent during follow-up were excluded.

| Indications for pacemaker implantation
Guideline-directed indications for pacemaker implantation were followed. 7 The choice of LP versus TVP was operator discretion.
At author's institution, the use of LP has steadily increased over time for single ventricular pacing. LP is particularly preferred in elderly patients and those with multiple comorbidities, chronic kidney disease, and dialysis. The majority of the patients in the TVP group had dual-chamber devices and underwent AV node ablation for rate control in the setting of permanent AF.

| Implantation procedure
Implantation of LP and TVP, as well as AV node ablation, was performed according to the standard technique described previously. 8,9 AV node ablation was performed either concurrently or during follow-up. Briefly, all patients underwent an ipsilateral venogram for the TV group before the axillary or subclavian vein access. The site of RV lead implantation was based on the discretion of the operator. For the LP group, a femoral approach was used for venous access. Deployment location was assessed in both right anterior oblique and left anterior oblique views. A chest X-ray was performed 2-4 h after the procedure in both groups to assess for pneumothorax (TV group) and lead/LP location. For patients undergoing concurrent AV node ablation, the devices were programmed for a lower rate of 80 beats per minute in a VVIR mode. All patients were seen in the device clinic after 2-4 weeks to assess pectoral incision, femoral access, and pacing parameters. For patients with AV node ablation, the lower rate was programmed to 60 or 70 based on operator preference.   Continuous data are represented as means and SDs. Student's t-test was used to compare continuous variables and χ 2 test was used to compare categorical variables with α = .05 for these tests.

| Data collection
Predictors of PICM were assessed in a stepwise manner: univariate analysis was first performed to identify candidate predictors; candidate predictors with p < .20 were promoted into a multivariate analysis and were incorporated into the final significance if p < .05.  Table 1. These characteristics were somewhat similar with some differences. Notably, fewer female patients received LP compared with TVP. The prevalence of CAD was higher in the TVP group.

| Implant and pacing characteristics
All patients in the LP and TVP groups were successfully implanted with pacemakers without any acute device-related complication.
Indication of pacemaker implantation, duration of QRS with and without pacing, and implant location are given in Table 2. Mainly, permanent AF with rapid ventricular response was the most common indication of pacemaker implantation in both groups. The majority of the patients in both groups had concurrent AV node ablation.
Patients in the LP group who did not have AV node ablation were also completely pacemaker dependent. The duration of QRS at baseline before pacemaker implantation was significantly longer in the LP group than TVP (115 ± 37 vs. 96 ± 22; p = .0001). Similarly, the duration of QRS after pacing was also significantly longer with LP (164 ± 24 vs. 155 ± 25; p = .01). The majority of the LP were implanted in the mid-septal (58%) or apical septal (31%) location.
The majority of the TVP was implanted in the apex (67%) followed by the apical septum (18%; Table 2).

| Primary outcome
Mean follow-up duration after AV node ablation in TVP and LP groups was 592 ± 549 and 817 ± 600 days, respectively. Overall, 18

| Acute and long-term procedure-related complications
Incidence of acute periprocedural and follow-up complications (>30 days) was similar in both groups as outlined in Table 3. Notably, one patient in the LP group had pericardial effusion noted immediately after deployment. A total of 45 (67%) patients in LP and 86 (67%) in the TVP group had an interpretable echocardiogram within 6 months of pacemaker implantation. At least one degree of TR worsening was seen in 34% in LP and 44% patients with TVP (p = .22; Table 3). Note: Data are given as n (%).

| Baseline ventricular function
Previous studies have identified various clinical characteristics as predictors of PICM. Baseline ventricular function before implantation of pacemaker has been identified as one risk factor for the development of PICM. This finding is the reason for the current recommendation of implantation of CRT in patients with ventricular dysfunction who are expected to require at least 40% pacing after implantation of a pacemaker. 3,4 This study aimed to identify the differences in LP and TVP for the development of PICM and thus excluded patients with ventricular dysfunction. The findings of this study clearly demonstrate that the incidence of PICM in patients with LP was significantly lower compared with TVP when baseline ventricular function is normal.

| Age
Age was a negative independent predictor for PICM on a multivariable analysis in this study. A small number of PICM patients in the LP cohort did not allow separate assessment of predictors in the LP cohort. These data may suggest that TVP and LP may be comparable for the development of PICM in elderly patients.

| Location of LP or pacing lead
This study clearly showed the differences in implantation  19 Majority of the patients who underwent CRT after PICM in this study noted an improvement in LVEF. The only two patients in the LP group who underwent CRT also showed significant improvement in LVEF. In both cases, the LP was abandoned, and a new RV and CS lead was implanted. Although abandoning an LP is an acceptable strategy, retrieval can also be attempted. 9

| Future perspectives
Before the availability of Micra AV, the use of LP was limited to patients who need single ventricle pacing, predominantly patients with infrequent need of pacing and those with permanent AF. Due to recent studies demonstrating the safety of LP over TVP due to the lack of lead and pocket-related complications, and availability of AV synchronous pacing, it is expected that the use of LP will increase over time. The findings of this study complement the previous studies and highlight another avenue where LP has fared better than TVP.

| Limitations
This study has some limitations. (i) The study is a retrospective single-center study from a high-volume center and the findings may not be evident in low-volume centers. (ii) The study is nonrandomized and device selection was based on operator discretion. Despite all these limitations, this is the first study suggesting the superiority of LP compared with TVP for the incidence of PICM and can serve as hypothesis-generating for larger future studies.

| CONCLUSIONS
This is the first study showing that the incidence of PICM is significantly lower for LP compared with TVP in pacemaker patients with normal ventricular function. Further prospective randomized trial is needed to validate these findings.