A case of leadless-to-leadless pacemaker interaction

Since the 2016 FDA approval of the Micra Transcatheter Pacing System (Medtronic Inc, Minneapolis, MN), the use of leadless pacemakers (LPMs) in the United States has grown substantially. In just over 2 years following approval, the volume of LPM procedures across all hospital types increased more than 6-fold, equating to an average quarterly increase of 24%. LPMs are particularly favorable for patients with high infection risk or compromised vascular access owing to the absence of both a subcutaneous generator pocket and transvenous leads. The safety and efficacy of leadless systems were initially supported by large multicenter clinical trials such as LEADLESS II and the Micra Transcatheter Pacing study and continue to be serially demonstrated by subsequent follow-up studies to date.^{,  ,  ,}  Furthermore, comparative outcomes from ongoing longitudinal studies are starting to suggest LPMs require less reintervention and have fewer complications than transvenous systems. Despite promising potential, the expanding utilization of leadless technology will undoubtedly be accompanied by novel challenges. The interaction and impact of multiple leadless devices is one arena warranting further characterization. We present a case of a patient who underwent reimplantation of a second LPM and subsequently required impromptu retrieval of his first Micra AV owing to intraprocedural development of interference between the 2 adjacent devices.

A 62-year-old veteran with end-stage renal disease on hemodialysis via right internal jugular tunneled catheter, severe aortic stenosis with normal left ventricular function (ejection fraction 60%–65%), and type 2 diabetes mellitus presented to our hospital with symptomatic complete heart block with a slow ventricular escape rhythm in the setting of hyperkalemia (6.2 mEq/L) from a missed dialysis session the day before. Transcutaneous pacing was performed in the emergency department owing to heart rate of 20 beats/min, until an active fixation temporary transvenous pacing lead was placed via his right internal jugular vein. Despite correction of his electrolyte derangements with dialysis, he continued to require intermittent pacing for third-degree atrioventricular (AV) block necessitating permanent pacemaker placement. Because of his poor transvenous options (right internal jugular tunneled dialysis catheter, immature left upper extremity arteriovenous fistula), he underwent successful implantation of an LPM (Micra AV). Sensing (20 mV) and pacing (0.63 V at 0.24 ms, 830 ohms) parameters on postprocedure day 1 were appropriate and he was discharged home with plans for outpatient follow-up.

Unfortunately, the patient re-presented 43 days later with recurrent symptomatic high-grade AV block with failure of Micra capture and intermittent atrial tracking (Figure 1). Device interrogation at that time revealed elevated thresholds (4.5 V at 0.4 ms) concerning for microdislodgement. Given ongoing relative contraindications to a transvenous system and increased long-term infection risk (end-stage renal disease with planned renal transplant), he underwent reimplant of a second Micra AV device. During the procedure, device placement away from the original Micra was unsuccessful; multiple deployment attempts were made at varying septal locations with poor thresholds. Finally, excellent parameters were verified in an area adjacent (though slightly more basal) to the initial device, and the second device was successfully placed at that location (Figure 2A and 2B , Supplemental Videos 1 and 2). Original device retrieval was considered at this point; however, we were concerned this would cause microdislodgement of the new Micra that was very challenging to place. There was initially no interaction for at least 10 minutes of observation. Just before Micra introducer sheath removal, a constant noise from device-device interaction was noted. Numerous ventricular oversensing events that occurred after ventricular pacing were identified on the intracardiac electrogram, as demonstrated in the representative tracing in Figure 3. This was most suggestive of mechanical interference between the 2 devices (the initial device was in nonpacing mode, “OFF”). This development prompted retrieval of the older device. In preparation, a transfemoral electrophysiology catheter was placed for additional pacing support as the patient remained intermittently pacer dependent. A 25mm Gooseneck snare through a medium-curve Agilis sheath was advanced into a 14F short sheath within the Micra introducer to prevent back bleeding; the Gooseneck straight catheter was then replaced with a Mach-1 multipurpose catheter to direct the snare more precisely. The Micra was securely snared via the retrieval knob, allowing a successful retrieval of the older device (Figure 2C, Supplemental Videos 3 and 4). The patient tolerated the complex case well, and his new leadless device continued to maintain normal sensing, impedance, and threshold parameters post explant of old device and on serial follow-up for the subsequent 20 months.

Device-device interactions (DDI) within conventional transvenous systems are a known, well-characterized phenomenon. To date, there is little data on DDI for leadless systems. Several case studies have described electromagnetic interference between left ventricular assist devices and Micra pacemakers, 1 of which required repositioning of the Micra to a location further from the LVAD casing to reestablish telemetry communication.^, To our knowledge, this is the first report of DDI between 2 LPMs.

Upfront retrieval of the older device had been considered at the start of the case. However, in hopes to save additional access and temporary pacing catheter placement (intermittent pacemaker dependence), placement of the new Micra was attempted first. Given the excessive number of device placement attempts required before deployment (13 contrast cines recorded), most frequently owing to unacceptable electrical parameters after deployment, we were concerned a subsequent retrieval attempt of the older device could result in microdislodgement of the newly placed device (which finally had acceptable parameters). No mechanical interaction occurred between the 2 Micra devices following deployment, tether removal, and observation, initially leading us to believe that abandonment of the old device was a feasible, and perhaps more favorable, option at that stage in the case, despite adjacent devices. Ultimately, we were proven otherwise when mechanical DDI declared itself later than anticipated.

As we enter a new era of innovate pacemaker technology, this case offers valuable insight into potential complications posed by multiple leadless devices. Implantation of additional devices may be required when an existing device is malfunctioning or at end of service. In our patient, significantly elevated thresholds (4.5 V at 0.4 ms) in the absence of a physiologic driver were indicative of suboptimal device positioning necessitating reintervention. Like transvenous systems, some degree of capture threshold elevation is tolerable immediately after Micra implantation and is expected to decrease over time with maturation of the device–tissue interface. However, capture thresholds greater than 2.0 V, as demonstrated in this case, are unlikely to normalize and warrant subsequent revision. Existing data suggest that the rate of system revision required 2 years from implantation for leadless pacing systems is 75% lower than that required for transvenous pacemakers. While this is encouraging, it is also notable that the current method for most leadless revisions is to disable the existing LPM and leave it affixed to the ventricular myocardium. According to cadaveric models, an average human right ventricle can accommodate 3 or more Micra devices. Given the anatomic feasibility, leaving an original device in situ may seem preferable when considering the risks of removal, particularly in pacemakers with longer dwell times. Although the expectation is that full encapsulation of a Micra will eventually occur, the endothelialization rate is unknown and likely variable.^, To our knowledge, the oldest Micra to be successfully retrieved percutaneously to date had a dwell time of 4 years. Although factors predicting high extraction difficulty require prudent consideration, as this case demonstrates, the potential for device–device interaction also needs to be considered when deciding between the elective removal or abandonment of LPMs. Patients whose myocardial substrate prevents adequate distancing between devices may be at higher risk of developing complications from electromagnetic interference or physical interaction. Until methods for characterizing the degree of encapsulation are proposed and a predictive timeline for ineffective percutaneous retrieval identified, LPM extraction should perhaps be more strongly considered. Admittedly, a paucity of data on many of these factors presently preclude a fair deliberation.

As leadless technology advances, indications for the implantation of multiple devices will extend far beyond basic device replacement. Clinical trials for a cardiac resynchronization system that uses leadless left ventricular pacing (WiSE-CRT) are currently underway and offer a window into the future. Integrated wireless systems involving various heart chambers and necessitating multidevice communication are on the horizon. For such developments to be feasibly implemented, leadless DDI needs to be better understood and defined. The characterization of DDI will allow us to enhance the selectivity of leadless devices, mitigate mechanical and electromagnetic interaction potential, and ensure safer, more effective pacing therapy.

Device–device interaction may be a significant limitation to multiple LPM device implantations and is an important consideration in periprocedural planning and future developments.