OUP user menu

Passive electrode effect reduces defibrillation threshold in bi-filament middle cardiac vein defibrillation

John R. Paisey, Arthur M. Yue, Frederick Bessoule, Paul R. Roberts, John M. Morgan
DOI: http://dx.doi.org/10.1093/europace/euj034 113-117 First published online: 11 January 2006

Abstract

Aims To investigate whether a passive electrode effect decreases defibrillation threshold (DFT) in multi-filament middle cardiac vein (MCV) defibrillation.

Methods and results Twelve pigs underwent active housing (AH) insertion, with defibrillation coils placed transvenously in right ventricular apex and superior vena cava. The MCV was cannulated, and 1.12F, 50 mm coil electrodes (Ela Medical SA, France) were deployed in its right and left branches. Lead placement was possible in 11 of 12 animals. DFT (J, mean±SD) was determined by three-reversal binary search and compared between the MCV monofilament (single filament deployed) and the AH (25.9±10.9) and the MCV mono+passive filaments (both filaments deployed, one connected) and the AH (19.9±11.4); 24% DFT reduction P=0.008.

Conclusion A bystander electrode adjacent to a monofilament electrode in the MCV reduces DFT by 24% when compared with monofilament MCV alone. Microfilament electrodes decrease DFT as auxiliary anode but not as sole anode.

  • Defibrillation threshold
  • Passive electrode
  • Cardiac vein
  • Bystander electrode
  • Implantable defibrillator

Introduction

The passive electrode effect is the influence of an electrode that is not connected to a circuit (a bystander electrode) on that configuration's defibrillation characteristics. It has been demonstrated that bystander epicardial patch electrodes increase defibrillation threshold (DFT), through a passive electrode effect, when shocking from transvenous systems.1,2 This effect does not appear to be replicated in bystander endocardial leads,3 and it has never been shown that the passive electrode effect may be used to decrease DFT.

Implantable cardioverter defibrillators (ICD) are an accepted treatment for primary46 and secondary79 prevention of life-threatening ventricular arrhythmias.

Evaluation of factors leading to a decrease in DFT may yield advantages in the clinical application of ICDs: failure rate will be reduced through an increase in safety margin (the difference between DFT and maximum output of device), and device size and longevity will be improved through the benefits of a lower DFT on battery and capacitor design.

The middle cardiac vein (MCV) has the potential to offer low DFT when compared with conventional endocardial systems through its anatomical location.10 It has been evaluated in animal1115 and human16 studies.

Methods

Aims

To investigate whether a passive electrode effect may decrease DFT in multi-filament MCV defibrillation using a novel microfilament electrode.

Ethical considerations

Approval was granted by the local regional Ethics Committee, and the British Government Home Office licensed the project and personnel.

Animal preparation

Twelve female pigs (weight 53.1±10.0 kg) were sedated with intramuscular benzodiazepines (Streznil® 0.2 mL/kg). After 15–45 min, general anaesthesia was induced with intravenous Saffan® (0.15 mL/kg) and the animals intubated. Anaesthesia was maintained with inhalational isoflurane (2% via oxygen at 10 L/min), and buprenorphine (0.2–0.4 mg) was given as adjuvant analgesia. Cut down was performed to the right internal jugular vein to facilitate transvenous lead placement. Intravenous normal saline was infused at 75 mL/h. The surface ECG was monitored continuously on lead II. Systemic blood pressure was monitored through a femoral artery cannula.

Defibrillation system configuration

A dual coil defibrillation lead (Sprint Quattro®, Medtronic, Minneapolis, MN, USA) advanced to the right ventricular (RV) apex. Thus, the coils were sited in RV and superior vena cava (SVC). The MCV was catheterized with an 8F MPA1 catheter cut to 58 cm. Custom-designed microfilament electrodes (1.12F with 50 mm length, 58 mm2 surface area coils, ELA medical SA, Montrouge, France), Fig. 1, were introduced into the left and right branches of the MCV. An active housing (AH) was inserted subcutaneously in the left pectoral area (Defender®, ELA medical). Electrodes were connected through a junction box to an external defibrillator (5358, Medtronic). Induction of VF was by 5 s 50 Hz AC current application, and defibrillation attempts were performed with a biphasic waveform with capacitive tilt.

Figure 1

Microfilament lead design.

DFT determination

DFT determinations were performed in two stages: the point of entry onto the final pathway was determined (Fig. 2) and the DFT was determined by a three-reversal binary search with increments determined by the programmability of the device (1 J increments up to 16 J and 2 J increments up to 18–34 J). If an animal could not be defibrillated by a certain configuration, the DFT was considered to be 34 J (maximum output of device) for analysis.

Figure 2

Determination of entry point onto DFT determination pathway. Therapies in joules. S, successful and F, failed defibrillation attempts.

Statistical analysis

Two configurations were compared to examine the passive electrode effect, a comparison between MCV (mono)→AH and MCV (mono+passive)→AH, Fig. 3.

In studies of DFT, it is conventional to compare values by a paired t-test. In this protocol, there were several instances of 34 J DFT values being allocated, because animals were not successfully defibrillated in the configuration concerned. The DFTs were not therefore normally distributed making parametric testing inappropriate. For this reason, the more rigorous non-parametric Wilcoxon signed rank test was used to assess significance. A two-sided P-value of 0.05 or less was considered significant.

Results

Placement of a bi-filament electrode was possible in 11 of 12 animals. In one animal, the MCV could not be selectively catheterized. Screening time for the procedure was 13.4±6.4 min. The DFTs and impedances of the configurations (Table 1 and Fig. 4) are shown.

Figure 4

DFT MCV monofilament to AH vs. MCV monofilament+passive to AH.

View this table:
Table 1

Summary results by configuration

AnodeCathodeDFT (J)±SDImpedance (ohm)±SD
MCV monoAH25.9±10.983.8±34.8
MCV mono+passiveAH19.9±11.472.5±26.9

DFT was 24% less in monofilament+passive than in mono alone to AH, P=0.008. The electrical properties of the microfilaments were atypical: at high energy outputs, their impedance increased substantially, preventing efficacious defibrillation in some animals for configurations involving microfilaments as sole cathode (Fig. 5).

Figure 5

Impedance vs. DFT for MCV sole anodal configurations.

Autopsy was performed at the end of the procedure. No macroscopic damage was seen in myocardium or pericardium.

Discussion

The anatomical site of electrodes alters the shocking vector and affects the DFT either by allowing inclusion of a critical mass of myocardium,17 of which the septum is an important region,10 or by permitting simultaneous depolarization of all the myocardium with sufficient energy to prevent recurrence of fibrillation.18,19 A higher DFT will result if energy is distributed unfavourably; the converse is a lower DFT with optimal electrode placement.10,20

Defibrillation configurations involving epicardial patch electrodes were superseded by transvenous systems because of the lower complication rate of the latter.21 This came at the cost of an increased DFT.20 An electrode placed transvenously but having the low DFT of the epicardial patch electrodes would combine the advantages of epicardial patch electrodes and transvenous systems.

Possible reasons for the lower DFT seen with epicardial patches are anatomical site (infero-septal), epicardial location, large electrode surface area, and broad area of myocardium in contact with the distal electrode.22

Defibrillation electrode placement in the MCV may decrease DFT13 because of its infero-septal epicardial location. Placement of multiple defibrillation filaments is feasible14 and gives further theoretical advantages by increasing the surface area of the electrode and broadening the amount of myocardium in contact with the distal electrode. Placement of multiple filaments increases complexity of implantation partly by necessitating multiple proximal connections: the passive electrode effect might be used to reduce this complexity.

After observations that bystander epicardial patches increase DFT through a passive electrode effect, but no equivalent influence on DFT is exerted by transvenous shocking coils or pacing leads, the phenomenon received no further research attention. We have shown that the passive electrode effect exerted by a bystander MCV coil in the radicle adjacent to an identical active coil decreases DFT by 24%.

The ability of a bystander electrode to exert a passive electrode effect is dependent on the proportion of current that is drawn through the alternative route. This is a function of the impedance of the intended configuration when compared with the impedance of the parallel circuit created by the bystander electrode and intervening tissue. Epicardial patches have lower impedance than transvenous coils, allowing current shunting and a passive electrode effect: two transvenous coils have similar impedance, minimizing current shunting. Furthermore, an electrical passive electrode effect may only influence DFT if it significantly alters shocking vector; in the case of bystander epicardial systems, current is drawn in the opposite direction from the anode: predictably DFT is increased. In transvenous systems, the intended and bystander coils occupy similar anatomical sites: the vector is not greatly altered even if a passive electrode effect exists on electrical properties.

The impedance characteristics of the microfilament electrodes used, Fig. 5, were unusual in that they were high and rose further with increasing current. This creates a situation where the actively connected electrode has a higher impedance than the bystander, favouring current shunting and a passive electrode effect. The placement of the two filaments, in the branches of the MCV adjacent to the septum, caused any current shunted to be to a site that is likely to improve current distribution (a virtual composite electrode, the connected and passive electrode, with a greater surface area and broader area of myocardium involved).

Placement of multiple filaments in the MCV radicles allows close mimicry of epicardial patch electrode: the anatomical equivalence of MCV with multiple filaments simulating the structure and current distribution. Using the passive electrode effect to avoid multiple proximal connections reduces the complexity of such a system.

For the passive electrode effect to be a clinically useful phenomenon, lead configurations taking advantage of it must be safe, stable, and transvenously deployed. It would be required either to reduce DFT by over 50% or significantly reduce the variability of DFT.

We have shown that the branching structure of coronary sinus tributaries, already widely used in pacing and validated in acute defibrillation studies,16 is a suitable site to explore the uses of the passive electrode effect in transvenous defibrillation. The electrical properties of the microfilaments made them unsuitable as sole anode as their impedance increased preventing effective defibrillation in certain individuals.

Limitations of the study

Two bystander electrodes present in this study were not examined for a passive electrode effect; the RV and SVC coils were left in place for configurations that did not involve them. They are less likely to exert a passive electrode effect given their distance from the active circuit and were constant in the MCV mono and mono+passive configurations.

The magnitude of the DFT reduction would not be clinically useful, and the configuration used was demonstrated not to be reliable for the reasons discussed.

It is not certain that findings from any animal study can be replicated in humans: discordance in findings between prior porcine and human MCV defibrillation studies13,15,16 (possibly explained by the anatomical difference in the mediastinal orientation between the species) has been seen.

In this case, however, the model, anatomical site, and magnitude of effect are secondary to the proof of concept: a passive electrode effect may reduce DFT.

Conclusion

In a porcine model, with transvenously placed coronary venous leads, a passive electrode effect decreases DFT and impedance when shocking to an AH.

Acknowledgement

J.R.P. receives research funding from Ela Medical SA, Montrouge, France.

References

View Abstract