© 2005 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.
SHORT SERIES REPORT
Left atrial far-field sensing by left ventricular leads: A potential hazard in cardiac resynchronisation therapy
aService de Cardiologie et des Maladies Vasculaires, Centre Hospitalier du Pays d'Aix Avenue des Tamaris, Aix-en-Provence, 13616 Cedex 1, France; bKarolinska Hospital Stockholm, Sweden; c Aarhus – Denmark; dMedtronic, Inc. France
Manuscript submitted 15 September 2003. Accepted after revision 1 July 2005.
*Corresponding author. Tel.: +33 442335053; fax: +33 442335169. E-mail address: jtaieb{at}ch-aix.fr
Abstract
BACKGROUND: Cardiac resynchronisation therapy (CRT) requires a lead advanced through the coronary sinus (CS) to pace the left ventricle (LV). Left atrial far-field signals (LAFFS) may be sensed by the LV lead at the time of implant or after lead dislodgement, and may inhibit ventricular pacing.
OBJECTIVE: To assess the incidence of detection of LAFFS > 2 mV and its correlation with the CS lead position.
METHODS: Data from the first 75 consecutive patients enrolled in the InSync III multicentre study were analysed. The position of the LV lead was recorded at implant. During follow-up, pacing was temporarily inhibited and the LV channel electrogram was recorded. The amplitude of LAFFS observed before discharge from the hospital and at 1 month of follow-up was retrospectively analysed. A LAFFS > 2 mV was considered clinically significant.
RESULTS: CRT systems were successfully implanted in 71 of 75 patients. A LAFFS > 2 mV was recorded by the LV lead channel in six of 71 patients (8.5%). This phenomenon developed between hospital discharge and 1 month of follow-up in two of these patients and in one case disappeared within 1 month. It was observed in all CS tributaries except the anterior and mid-cardiac veins.
CONCLUSIONS: Left atrial far-field signals sensed by the LV lead were not rare. Implanting physicians should be aware of this phenomenon in order to prevent potentially serious complications.
Key Words: left atrial far-field signal, atrial far-field sensing, cardiac resynchronisation therapy, biventricular pacing
Introduction
The clinical benefits offered by cardiac resynchronisation therapy (CRT) have been recently confirmed by randomised, controlled trials[1,
2]. Depending on its location, the lead placed in the coronary sinus (CS) venous system to stimulate the left ventricle (LV) may sense signals originating from the left atrium and inhibit biventricular stimulation. This phenomenon may be responsible for absence of resynchronisation or asystole if the patient is pacemaker dependent[3–6]. The incidence of this undesirable phenomenon, and its relationship with the site of LV lead implantation has not been reported. We present a retrospective analysis of a consecutive series of patients enrolled in a multicentre study of CRT, with special attention to left atrial (LA) far-field signals (FFS) recorded by the LV lead after pacemaker implantation and 1 month of follow-up.
Methods
The patients included in this analysis had been enrolled between October 2000 and May 2001 in INSYNC 3, a multicentre study of the INSYNC 8042 triple chamber pulse generator (Medtronic Inc., Minneapolis, MN, USA). The device, which includes three separate pacing/sensing channels, was implanted in patients suffering from congestive heart failure and spontaneous or right ventricular (RV) pacing-induced intraventricular conduction abnormalities. The LV leads successfully implanted in this study included Medtronic models 2187 (n = 48), 4189 (n = 16), and 4191 (n = 7).
All patients underwent temporary and separate testing of each pacing/sensing channel before being discharged from the hospital and at 1 month of follow-up. Signals sensed by the LV channel were recorded during spontaneous rhythm (pacing OFF). The signals were telemetered from the implanted pulse generator and manual pick–pick measurements were made with a ruler from the programmer's printed tracings, recorded in real-time at the time of device interrogation. This retrospective analysis included tracings obtained from the first 75 patients enrolled in this study. A cut-off value of 
2 mV for LAFFS was based on the potential risk of pacing inhibition if the device is left programmed at nominal values, which, in currently commercially available systems, vary between 2.0 and 2.8 mV.
Results
Reliable biventricular stimulation was achieved in 71 of the 75 patients. The lateral and the postero-lateral veins were chosen as sites of LV lead implantation in 54 of the 71 patients (76%). In the 17 other cases these suitable veins could not be reached and the lead was placed in another CS tributary to pace the left ventricle. LAFFS were observed in six of the 71 patients (8.5%), before hospital discharge in four, and at 1 month of follow-up in two other patients. It disappeared in one case within 1 month (Tables 1 and 2). LAFFS did not occur in patients with leads placed in the anterior or mid-cardiac veins. No adverse effect was observed that was attributable to ventricular pacing inhibition. No patient was pacemaker dependent. No asystole was observed during the pacing inhibition test either at predischarge, or at 1 month.
|
|
Discussion
Mechanisms and consequences of LAFFS sensing
In patients with sinus rhythm, since dedicated CRT systems with separate ventricular ports are not widely available in certain parts of the world (for instance, they have not been approved for use by all private medical centres in France in 2005), "triple chamber" systems often consist of a standard dual chamber pacemaker with a right atrial lead conventionally connected to the atrial channel, and RV and LV leads both connected to the RV channel via a Y connector, with the LV lead typically placed in a lateral CS tributary[7]. The signal sensed by the ventricular channel is, thus, biventricular (Fig. 1). The placement of the LV lead is associated, in certain locations, with a risk of LAFFS sensing (Fig. 2A and B), which may be apparent at the time of CRT implant, or develop after lead dislodgement, as was observed in two of our patients. In one patient, LAFFS disappeared within 1 month. Variations in the anatomy of the CS tributaries may limit the choice of LV lead placement[8,9], and a compromise may have to be made among LV signal amplitude and delay[7], sensing of an LA electrogram, and capture threshold at that site. In addition, particularly with unipolar CS leads, diaphragmatic stimulation, by mandating the withdrawal of the lead to a more proximal site where a left atrial far-field signal is recorded, may force the abandonment of an otherwise suitable site. In the pursuit of a delayed LV signal, currently considered one of the important goals in the proper placement of the LV lead[7], one may, therefore, have to accept some degree of LAFFS sensing (Fig. 3). However, if the signal amplitude is larger than the biventricular channel sensing threshold, stimulation is inhibited[3–5]. This is not only a source of CRT failure but, in pacemaker-dependency, a cause of asystole (Fig. 4)[4]. Finally, inappropriate discharges of combined CRT/ICD systems caused by double counting of LAFFS sensing have also been reported[6].
|
|
|
|
Extrapolation of left ventricular lead far-field sensing to CRT systems with Y adaptors
Far-field left atrial signals, when present, are always detected before the right and the left ventricular EGM during sinus rhythm, although they are sensed by the left ventricular lead. Therefore, when left atrial far-field signals are present, they are recorded at precisely the same time and at the same amplitude, whether by a biventricular sensing lead (Y adapter) or by a left ventricular sensing lead. This is why we felt that it was appropriate to extrapolate the observations (and their potential consequences) of left ventricular lead only recording in separate channel devices, to the far-field sensing of left atrial signals by CRT systems with Y adaptors.
Incidence of LAFFS sensing
Depending on the device model implanted, the nominal ventricular sensing setting of the biventricular channel varies between 2 and 2.8 mV. Therefore, unless reprogramming to a lower sensitivity has been performed, sensing of LAFFS may occur if LA signals are >2 mV. In this study, LAFFS > 2 mV were recorded in 8.5% of patients who had undergone successful implantation of CRT systems and, in 3%, they appeared after discharge from the hospital, and in one patient, disappeared within the first month. Though LAFFS were not recorded in the patients whose LV leads were positioned in the anterior and mid-cardiac veins, their presence should be expected whenever the lead is situated near the mitral annulus, i.e. in a CS tributary close to the left atrium. Failure to detect this phenomenon may have serious short- and long-term consequences if not anticipated and prevented.
While proficient operators may have access to a wide choice of LV lead implantation sites, it is noteworthy that the participants in this multicentre study were experienced investigators, suggesting that our results are reflective of actual clinical practice.
Proposed solutions
The detection of stimulation inhibition by LAFFS sensing is a manoeuvre of particular importance, which can be performed intra- and/or postoperatively:Intraoperative testing: During implantation of the LV lead, the LV electrogram should be continuously recorded, measured manually, while meticulously searching for a stable site of lead implantation away from the mitral annulus.
Postoperative testing: In the presence of a persistent LA electrogram after implantation of the lead, several measures can be taken to prevent the occurrence of biventricular stimulation inhibition:
- Stable right atrial pacing by setting a high back-up rate or by implementing an atrial overdrive algorithm. During right atrial pacing, sensing of an LA signal is then interpreted as atrio-ventricular (AV) cross-talk, initiating delivery of pacing in the safety window. The disadvantage of this form of protection is the resulting AV delay, which may differ significantly from the value chosen during post-operative echo-Doppler examination to optimise CRT.
- Programming of a biventricular channel sensitivity far lower than the nominal value. This creates a risk of undersensing of ventricular ectopic activity, pacing in the ventricular vulnerable period, and could trigger ventricular tachyarrhythmias.
- Sensing of ventricular activity confined to the right ventricle, which eliminates the risk of LAFFS sensing[10]. This non-operator-dependent option, built-in at the time of device manufacture, is distinctly the safest. In this study, all pacemakers were programmed in the nominal RV only sensing configuration, which explains the absence of adverse effects related to ventricular pacing inhibition in this group of patients.
In conclusion, LA far-field sensing by the LV lead of CRT systems, though rare, may be encountered in daily practice, even when implanted by experienced operators. Awareness and anticipation of this phenomenon is urged, and the option of isolated RV sensing should be made available and programmed in all CRT systems.
References
[1] Cazeau S, Leclercq C, Lavergne T, for the Multisite Stimulation In Cardiomyopathies (MUSTIC) Study Investigators 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.
[2] Abraham WT, Fisher WG, Smith AL, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346: 1845–1853.
[3] Lipchenca I, Garrigue S, Glikson M, Barold SS, Clémenty J. Inhibition of biventricular pacemakers by oversensing of far-field atrial depolarization. Pacing Clin Electrophysiol 2002; 25: 365–367.[Medline]
[4] Taieb J, Benchaa T, Foltzer E, et al. Atrioventricular cross-talk in biventricular pacing: a potential cause of ventricular standstill. Pacing Clin Electrophysiol 2002; 25: 929–935.[Medline]
[5] Oguz E, Akyol A, Okmen E. Inhibition of biventricular pacing by far-field left atrial activity sensing: case report. Pacing Clin Electrophysiol 2002; 25: 1517–1519.[Medline]
[6] Vollmann D, Lüthje L, Görtler G, Unterberg C. Inhibition of bradycardia pacing and detection of ventricular fibrillation due to far-field atrial sensing in a triple chamber implantable cardioverter defibrillator. Pacing Clin Electrophysiol 2002; 25: 1513–1516.[CrossRef][Medline]
[7] Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, Santini M. Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol 2002; 39: 489–499.
[8] Meisel E, Pfeiffer D, Engelmann L, et al. Investigation of coronary venous anatomy by retrograde venography in patients with malignant ventricular tachycardia. Circulation 2001; 104: 442–447.
[9] Gilard M, Mansourati J, Etienne Y, et al. Angiographic anatomy of the coronary sinus and its tributaries. Pacing Clin Electrophysiol 1998; 21: 2280–2284.[CrossRef][Medline]
[10] Taieb J, Benchaa T, Foltzer E, et al. Interest of configurable sensing biventricular pacemaker to prevent left atrial sensing by coronary sinus lead. J Am Coll Cardiol 2002; 39:Suppl. B 66B [Abstract].
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||