Left ventricular pacing in patients with preexisting tricuspid valve disease

Abstract Background Conventional right ventricular (RV) pacing is increasingly recognised to cause tricuspid valve (TV) injury or dysfunction, in part due to the need to pass the lead through the valve. This may be especially problematic in patients with preexisting TV disease or prior TV surgery. An alternative in this situation is to implant a left ventricular (LV) lead instead of ventricular pacing. Methods We performed a single‐center retrospective analysis of 26 patients with tricuspid valve surgery/disease who received a LV pacing lead in the coronary veins to avoid crossing the tricuspid valve, with or without a right atrial lead. A matched control population was obtained from patients receiving conventional right ventricular pacing and outcomes were compared. Main outcomes of interest were lead stability, electrical lead parameters and change in echocardiographic parameters such as left ventricular ejection fraction (LVEF) during long‐term follow‐up. Results Successful left ventricular pacing was established in 25 out of the 26 cases with one case converted to a RV lead due to lead dislodgement. During the 2.96 ± 1.0 year follow‐up, 24 of 25 (96.0%) leads were functional with stable pacing and sensing parameters, and 1 of 25 (4.0%) was extracted for due to device infection following an episode of thrombophlebitis. Conclusion We conclude that in patients with existing tricuspid valve disease or surgery, ventricular pacing via the coronary veins is a feasible, safe, and reliable alternative to right ventricular pacing.


| INTRODUC TI ON
Implantable pacemakers are widely used in the treatment of patients with bradyarrhythmias and the conventional approach has been to implant leads into the right ventricle (RV), most commonly at the RV apex. The RV apex is often the site of choice for ventricular pacing given the relative ease and safety of implantation, low risk of displacement and good reliability. 1 However, conventional RV pacing is increasingly recognised to cause tricuspid valve (TV) injury or dysfunction, in part due to the need to pass the lead through the valve. [2][3][4][5][6] This is thought to occur through several different mechanisms including direct mechanical trauma and damage to the valves or due to interference with the valvular mechanism and may be especially problematic in patients with preexisting TV disease or prior TV surgery. 7 Furthermore, long-term right ventricular pacing has been shown to be deleterious to left ventricular function in some patients as it induces left ventricular dyssynchrony and can result in heart failure. 8 There has been various experiences with patients in whom transvalvular leads are contraindicated, most prominently in patients with a mechanical tricuspid valve. In recent years there has been greater experience with LV pacing due to the advent of biventricular pacemakers or defibrillators and the use of contemporary left ventricular leads which are placed in the coronary veins in such devices. 9,10 In particular, the increasing use of cardiac resynchronization therapy device implants which involves LV lead implantation in the coronary veins has shown that LV pacing can be performed with low rates of lead dislodgement. 11 This suggests the potential utility of this approach in patients with existing tricuspid valve pathologies.
In this report, we reviewed our experience with patients with tricuspid valve disease or prior surgery who required pacemaker implantation in whom a standard LV pacing lead was used in place of a conventional RV lead to demonstrate the safety and feasibility of this approach. Outcomes were compared to a matched group of patients with a conventional RV lead.

| ME THODS
A retrospective review of 26 consecutive patients who underwent pacemaker implantation with an LV lead from December 2012 to December 2015 was conducted. These patients were undergoing pacemaker implantation for standard indications but also had significant tricuspid valve disease (defined as moderate or severe regurgitation), or previous tricuspid valve surgery. They were all implanted with a standard single or dual chamber pacemaker with a bipolar LV pacing lead in place of a conventional RV lead. Procedures were all performed under local anaesthesia and with light sedation.
The choice of the left ventricular lead and the vein targeted for implantation was left to the discretion of the operator. The patients were then followed as per usual clinical practice with pacemaker checks on 1 day, 1 week, 1 month, and 6 months after implantation. Subsequent follow-ups were at yearly intervals or at the physician's discretion. Post implant transthoracic echocardiography was also performed to monitor for any change in the left ventricular function and for valvular pathology.
A set of 1:1 matched controls were then obtained from the institution pacemaker implant database during the same time period (2012-2015) from patients receiving a conventional RV pacing lead.
The controls were obtained by matching within the database for several key parameters notably: (a) age (less than 3 years age difference at time of device implantation), (b) gender, (c) less than 10% difference in left ventricular ejection fraction (LVEF) as well as (d) the type of pacemaker implanted (single or dual chamber). The first case that The two groups were compared for multiple parameters that were monitored over the duration of follow-up where the primary outcome was lead stability over time. Categorical variables were compared using the chi-square or Fisher exact test. Within-group comparison was undertaken using Mann-Whitney tests for echocardiographic data or repeated measures ANOVA testing for electrophysiologic data. A P < .05 was considered the threshold for statistical significance for all analyses. All statistical analyses were performed using SPSS Version 22 (IBM Corp.).

| RE SULTS
A conventional bipolar LV lead was implanted as part of a dual chamber (n = 14) or single chamber (n = 12) pacing system. Patients were followed up for 1081 ± 365 days to determine lead performance.
A brief description of the clinical characteristics of the patients enrolled in the study is presented as Table 1. In terms of the tricuspid valve disease involved, most of the patients had moderate to severe organic TR (21) with 4 cases of 20 prosthetic valves involved (3 metallic valves and 1 bioprosthetic valve).
In all 26 cases, implantation was initially successful and the majority of the LV leads were implanted in the lateral cardiac vein. Two patients required lead adjustments on day 1 of implantation due to noted lead dislodgement. In the first case, the LV lead was originally implanted into the middle cardiac vein but dislodged. On reassessment, there were high thresholds in the anterior cardiac vein while in the more septal branches lead stability was questionable as the guidewire would freely enter the right ventricle. Finally, the LV lead was re-implanted into the middle cardiac vein successfully. In the second case, the LV lead was initially implanted into the posterolateral vein. On reassessment, the middle cardiac vein and great cardiac vein branches had poor thresholds and the anterolateral vein could not be cannulated successfully, after much attempts, decision was made to convert to a conventional RV lead due to the lack of a suitable landing site within the coronary veins. were also unable to be captured in the study. There was no significant difference in mortality or lead complications between the two groups. Table 2 shows the baseline and procedural characteristics of the case and controls. The controls were matched to the cases for age, gender, LVEF (within 10%) as well as the type of device implanted (single or dual chamber). The primary indications for pacemaker were sinus node dysfunction, complete heart block, slow AF with high grade AV block, and tachy-brady syndrome in both populations. It was shown that over time the lead impedance remained stable throughout duration of follow-up in both RV and LV leads (P = .196).
Over the long-term follow-up, the impedance in the LV leads remained higher than that in the RV leads (P < .001).
One significant observation was that the fluoroscopy times for the LV pacemakers were significantly longer than that for the RV pacemakers. However, it should be noted that excluding the first 3 cases which took a longer time perhaps due to the learning curve for the LV pacemaker implantation, the average fluoroscopy time for an LV pacemaker was 23min 32s.
Follow-up transthoracic echocardiography was performed routinely pre-device implantation and at around 2 years post implantation and findings are shown in Table 4. The LVEF in general

| D ISCUSS I ON
This report shows the overall safety and feasibility of the LV pacing approach in patients with tricuspid valve disease in whom a right frequently in patients with a mechanical tricuspid valve and traditionally an epicardial system was implanted. However, epicardial leads typically require sternotomy or thoracotomy for insertion and tend to have lower reliability and higher pacing thresholds than their endocardial counterparts. 9 With the advent of coronary sinus lead placement, permanent leads can now be delivered transvenously for the purpose of ventricular pacing in these patients without the need to cross the tricuspid valve. 10 However there remain concerns regarding the feasibility of ventricular pacing via the coronary sinus as complications can include diaphragmatic stimulation, coronary sinus dissection and lead displacement. 13 Particularly, due to anatomical considerations and a lack of active fixation with all CS leads bar one currently, lead stability had been a concern over the years.

| LV lead parameters remain stable
However, in this study we found that over 3 years' follow-up, there were no cases of lead malfunction or failure among our LV paced cases and only one case did develop MSSA bacteraemia requiring lead extraction. We also found that while LV pacing thresholds were higher than the conventional pacing thresholds as in RV pacing, in keeping with prior reports. 14,15 Overall, the LV capture thresholds remained stable and all were below 2.0 V. Furthermore, the relatively higher pacing impedance in LV leads would also mitigate their higher pacing thresholds in relation to battery drain and hence the impact on battery longevity may not be as significant. It is also noteworthy that we did not have any problems with diaphragmatic capture in this case series after the pacing parameters were optimized before discharge after device implantation.

| Longer procedural times
In this present series the average fluoroscopy time when implanting an LV lead was significantly longer than the corresponding RV pacemaker procedures. This reflects the additional steps of accessing the coronary sinus as well as the greater technical challenge of the procedure which is contributed by the variability in coronary venous anatomy and the lack of muscular trabeculae that aid in anchoring a lead in the RV. 14 In our experience, the first three cases had a significantly longer fluoroscopy time of longer than an hour which is expected given the initial learning curve. Excluding these cases, the average fluoroscopy time of implanting an LV lead was around 23 minutes and comparable to that reported by other studies as well. 15,16 However, the overall success rate in our experience does show that in trained hands, despite the relative difficulty of the procedure, the procedure remains one that is safe and feasible.
In terms of selecting the site for lead implantation, the mid-lateral wall has traditionally been the preferred site to maximize car-

| No change in LVEF
Over the course of the follow-up, there was no significant deterioration in the LVEF in patients with either an RV or LV lead. However, there was a non-significant numerical improvement in patients with an LV lead while those with an RV lead had a slight decrease. This is in keeping with previous studies that long-term right ventricular pacing may be deleterious to long-term cardiac function. 8

| Limitations
A small retrospective analysis like ours has inherent limitations in ascertaining cause and effect and this does apply to our study. A further limitation is that we were unable to determine the causes of death for the patients who passed on in the community and whose causes of death were not captured in the national electronic health records. For this study, we were unable to access the national death registry due to patient privacy laws. Moving forward, a prospective study evaluating LV versus RV pacing for routine pacemaker implants may be useful in confirming and further elaborating on the findings of this study.
With the availability of leadless pacemakers recently, it could be argued that LV pacing to prevent tricuspid valve issues is now less important. However, current leadless systems can only pace the ventricle and their whole life performance is still unknown. In younger patients who are likely to live longer than the expected battery longevity, the ability to and safety of implanting multiple leadless pacemakers is still not clear. Hence, there is still a role for LV pacing systems in such patients with tricuspid valve disease who require ventricular pacing, especially in the form of dual chamber pacemakers.

| CON CLUS ION
LV pacing with a lead through the coronary venous system is a safe alternative with good long-term reliability in patients who require pacing but who have contraindications to placement of a lead across the tricuspid valve.

CO N FLI C T O F I NTE R E S T S
The authors declare no conflict of interests for this article.