Performance of Dedicated Versus Integrated Bipolar Defibrillator Leads with CRT-Defibrillators: Results from a Prospective Multicenter Study

Transcription

1 Performance of Dedicated Versus Integrated Bipolar Defibrillator Leads with CRT-Defibrillators: Results from a Prospective Multicenter Study ROGER A. FREEDMAN, M.D.,* ALEX PETRAKIAN, M.D., KER BOYCE, M.D., CHARLES HAFFAJEE, M.D., JESUS E. VAL-MEJIAS, M.D., and ASHISH L. OZA, M.S.** From the *University of Utah Hospitals and Clinics, Salt Lake City, Utah; East Texas Medical Center, Tyler, Texas; First Health Moore Regional Hospital, Pinehurst, North Carolina; St. Elizabeth s Medical Center of Boston, Boston, Massachusetts; Galichia Heart Hospital, Wichita, Kansas; and **St. Jude Medical, Sylmar, California Introduction: Right ventricular (RV) anodal stimulation may occur in cardiac resynchronization therapy defibrillators (CRT-D) when left ventricular (LV) pacing is configured between the LV lead and an electrode on the RV defibrillator lead. RV defibrillator leads can have a dedicated proximal pacing ring electrode (dedicated bipolar) or utilize the distal shocking coil as the proximal pacing electrode (integrated bipolar). This study compares the performance of integrated versus dedicated leads with respect to anodal stimulation incidence, sensing, and inappropriate ventricular tachyarrhythmia detection in patients implanted with CRT-D. Methods: Two hundred ninety-two patients were randomly assigned to receive dedicated or integrated bipolar RV leads at the time of CRT-D implantation. Patients were followed for 6 months. Results: Patients with dedicated bipolar RV leads exhibited markedly higher rates of anodal stimulation than did patients with integrated leads. The incidence of anodal stimulation was 64% at implant for dedicated bipolar RV leads compared to 1% for integrated bipolar RV leads. The likelihood of anodal stimulation in patients with dedicated leads fell progressively during the 6-month follow-up (51.5%), but always exceeded the incidence of anodal stimulation in patients with integrated leads (5%). Clinically detectable undersensing and oversensing were very unusual and did not differ significantly between lead designs. There were no inappropriate ventricular tachyarrhythmia detections for either lead type. Conclusion: Integrated bipolar RV defibrillator leads had a significantly lower incidence of RV anodal stimulation when compared to dedicated bipolar RV defibrillation leads, with no clinically detectable oversensing or undersensing, and with no inappropriate ventricular tachyarrhythmia detections for either lead type. (PACE 2009; 32: ) integrated, dedicated, anodal stimulation, undersensing, oversensing, inappropriate therapies *Financial Support: This study was sponsored by St. Jude Medical. Conflict of Interest: (1) Dr. Roger Freedman, St. Jude Medical: research support, consulting, fellows program support; Boston Scientific: research support, consulting, fellows program support; Medtronic: research support, consulting, fellows program support; Sorin: consulting. (2) Dr. Alex Petrakian, St. Jude Medical: stock ownership; Boston Scientific: stock ownership; Medtronic: stock ownership. (3) Dr. Ker Boyce, St. Jude Medical: Research support, consulting; Boston Scientific research support; Spectranetics physician proctor. (4) Dr. Charles Haffajee, St. Jude Medical: Research support; Biotronik: research support; Boston Scientific: Research support, consulting; Medtronic: research support; ELA- Sorin: consulting, national PI. (5) Dr. Jesus E. Val-Mejias, St. Jude Medical: Research support, fellows program support. (6) Mr. Ashish Oza, St. Jude Medical CRMD: employee, stock ownership. Address for reprints: Roger A. Freedman, M.D., University of Utah Hospitals and Clinics, 30 North 1900 East, 4A100 UUHSC, Salt Lake City, UT Fax: ; roger.freedman@hsc.utah.edu Received July 22, 2008; revised September 14, 2008; accepted October 16, Introduction Cardiac resynchronization therapy (CRT) with biventricular pacing has been established as an effective treatment in selected patients with drug refractory heart failure. Various studies have demonstrated the clinical benefit of CRT with improvements in quality of life, heart failure symptoms, exercise capacity, mortality, and hospitalization rates. 1 3 In the United States, CRT is usually incorporated into devices that also have defibrillator capability given that this patient population has been shown to be at risk of sudden cardiac death. 4,5 An essential component of CRT defibrillators (CRT-D) is the right ventricular (RV) defibrillation lead. RV defibrillation leads are available with two configurations of sensing and pacing electrodes. (Fig. 1). Dedicated bipolar RV defibrillator leads have a ring electrode just proximal to the distal pacing electrode, similar to bipolar pacemaker leads, which is used as the anode for pacing and sensing. Integrated bipolar RV defibrillator leads do not have a proximal ring electrode; rather, the C 2009, The Authors. Journal compilation C 2009 Wiley Periodicals, Inc. PACE, Vol. 32 February

2 FREEDMAN, ET AL. Figure 1. Anodes for cross-chamber LV pacing using a dedicated bipolar RV lead (left) and an integrated bipolar RV lead (right). In both instances, the cathode for LV pacing is the LV pacing lead electrode. distal shocking coil is used as the anode for pacing and sensing. In CRT devices, left ventricular (LV) pacing is often done with the LV lead tip electrode serving as the cathode and the RV ring/coil as the anode. This configuration of pacing has been variously termed cross chamber pacing, extended bipolar pacing, and dual cathodal pacing. In most cases with this configuration, myocardial depolarization occurs only at the LV (cathode). However, in some patients myocardial depolarization may also occur from the RV ring/coil (anode) at sufficiently high output voltages; this is termed anodal stimulation. When anodal stimulation occurs, there is capture at both the RV and LV when only LV pacing was intended. If the CRT device is programmed to deliver the LV stimulus prior to the RV stimulus ( LV offset ) in order to improve resynchronization efficacy, anodal stimulation negates such offset since both the LV and RV will be depolarized simultaneously from the programmed LV stimulus. The subsequent RV stimulus, which typically would be delivered ms after the programmed LV stimulus, would encounter refractory myocardium and not result in RV cathodal capture. At very short V-V intervals (< 20 ms), cathodal stimulation from RV stimulus is still possible. 6 In either case, sequential V-V pacing is disrupted because of anodal stimulation resulting in unintended simultaneous biventricular stimulation. Anodal stimulation is also potentially arrhythmogenic. Ventricular refractoriness is shorter for anodal versus cathodal stimulation, particularly early after lead placement, and is associated with an increase in the vulnerable period for arrhythmias in the setting of anodal stimulation. 7 9 Additionally, anodal stimulation can make it difficult to determine LV pacing thresholds, occasionally resulting in an overestimation of the true LV pacing threshold and programming of unnecessarily high LV pacing outputs. On the other hand, anodal stimulation has been reported in some cases to have a beneficial effect on LV synchrony and LV systolic performance during biventricular pacing. 10 This prospective, randomized trial was designed to compare dedicated and integrated bipolar RV defibrillator leads in patients receiving CRT-D with respect to anodal RV stimulation, clinically detectable RV undersensing and oversensing, and inappropriate ventricular tachyarrhythmia detection due to oversensing. Methods Patient Population Nineteen centers enrolled and successfully implanted CRT-D devices in 292 patients. The centers and investigators are listed in the Appendix. The study was reviewed and approved by the appropriate Human Research Ethical Committees of each of the participating medical centers, and all patients gave written informed consent for participation in the study. Of the enrolled patients, 267 had not been previously implanted with a pacemaker or defibrillator. The other 25 patients were receiving a new RV defibrillator lead to upgrade or revise an existing pacemaker or defibrillator system. Patients were ineligible for the study if they had a mechanical valve in the tricuspid position, if their life expectancy was less than 6 months, if they were pregnant or planning pregnancy within 6 months, or if they were less than 18 years old. Lead System Implantation All patients were implanted with a Food and Drug Administration-approved CRT-D generator (Epic HF or Atlas HF models V-337, V-338, V-340, V343 or V355, St. Jude Medical, Sylmar, CA, USA). Patients were randomized and implanted with either a dedicated bipolar RV defibrillator lead (St. Jude Medical Riata or Riata ST models 1580, 1581, 1582, 7000, 7001, or 7002) or an integrated bipolar RV defibrillator lead (St. Jude Medical Riata I models 1590, 1591, 7010, or 7011) (Fig. 1). A variable randomization block scheme, within centers as the blocking factor, was 158 February 2009 PACE,Vol.32

3 INTEGRATED VS. DEDICATED DEFIBRILLATION LEADS used to provide the study center s randomization schedule. Both the dedicated and integrated RV leads were available in single- and dual-shocking coil designs and the choice between these was not controlled in this study. For both RV lead configurations, the surface area of the distal pacing electrode was 8 mm 2. The dedicated RV leads have a proximal pace/sense anodal ring electrode with surface area of 17 mm 2 and tip to ring spacing of 11 mm. The integrated RV leads do not have a proximal ring electrode and use the distal shocking coil as the pacing anode, with surface area of 367 mm 2 (models 7000, 7001, 7002, 7010, 7011) or 414 mm 2 (models 1580, 1581, 1582, 1590, 1591), all with a tip-to-coil spacing of 11 mm. Anodal Stimulation Threshold Testing and LV Capture Threshold Testing Anodal stimulation and LV capture threshold testing (Fig. 2) was performed by initiating LV pacing in the LV tip to RV ring/coil pacing configuration (cross chamber pacing) starting at 7.5 V and 0.5-ms pulse width and then progressively decrementing the output voltage. Surface electrocardiogram (ECG) and intracardiac electrograms were recorded during the testing. Anodal stimulation was identified by a local electrogram recorded on the RV lead at, or nearly at, the same time as LV pacing stimulus. Loss of anodal stimulation was identified by time displacement of the RV electrogram and a change in the surface QRS complex. Loss of LV capture was identified by the absence of a local LV electrogram and absence of a paced QRS complex on surface ECG. The anodal stimulation threshold was defined as the lowest LV output voltage resulting in anodal stimulation at 0.5-ms pulse width, and LV capture threshold was defined as the lowest LV output resulting in LV capture at 0.5-ms pulse width. Since pacing output is typically programmed to twice the capture threshold, patients who exhibited an anodal stimulation threshold less than or equal to twice the LV capture threshold were considered to have clinically relevant anodal stimulation. Device Programming Ventricular sensitivity was programmed using an automatic sensitivity control algorithm, with 93% of the patients programmed to the nominal maximum sensitivity value (0.3 mv). The ventricular tachyarrhythmia detection zones were programmed so that the lowest detection rate criterion was no higher than 150 beats/min. Stored electrogram recording was programmed to trigger on the diagnosis of an atrial or ventricular tachyarrhythmia. The duration of the stored electrogram prior to the trigger was set to 16 seconds (50%), 14 seconds (45%), or 12 seconds (5%). Follow-Up Patients were seen and their devices were interrogated and tested at intervals of 1, 3, and 6 months after implantation. Study follow-up was terminated 6 months postimplant. At the time of implant and at each follow-up, routine RV capture and sensing thresholds were measured. LV capture threshold and RV anodal stimulation threshold were measured with the cross-chamber pacing configuration as described above. At all follow-up visits, device-based stored electrograms were evaluated by each participating investigator to identify any episodes of at least one undersensed or oversensed beat or any inappropriate ventricular tachyarrhythmia detection. Inappropriate detection was defined as any detection interpreted by the device as a ventricular tachyarrhythmia Figure 2. LV capture threshold test in a patient with a dedicated bipolar RV lead, showing loss of anodal stimulation and loss of LV capture. Note the change in surface paced QRS morphology and the change in RV electrogram morphology when anodal stimulation is lost. PACE,Vol.32 February

4 FREEDMAN, ET AL. where the corresponding stored electrogram showed the presence of myopotential or other electromagnetic interference resulting in the detection. All episodes of undersensing or oversensing with documentation were reviewed by a single investigator (R.A.F.) for confirmation. Statistical Analysis A Fisher s exact test was used to compare the incidence of anodal stimulation, undersensing, and oversensing between the integrated and dedicated leads. A Cuzick s nonparametric test was used to check for trends in the incidence of clinically relevant anodal stimulation, undersensing, and oversensing over time across each group. A two-group t-test of equal means was used to compare RV pacing threshold, RV sensing threshold, LV pacing threshold, and LV pacing impedance between the dedicated and integrated leads. A p-value < 0.05 was considered statistically significant. All tests and fits were carried out in SAS (SAS Institute Inc., Cary, NC, USA) and Stata (Stata, Statistics/Data Analysis, College Station, TX, USA). Results Patient Demographics The study population of 292 patients was typical of those receiving CRT-D devices (Table I). Most of the patients were men, most had coronary artery disease as their underlying structural heart disease, and most were New York Heart Association (NYHA) functional class III. The typical LV ejection fraction was in the 20 30% range. Sixteen (5%) of the patients were in NYHA functional class II; patients with NYHA class II symptoms do not meet standard criteria for CRT implantation, and the decision to implant CRT-D systems in these patients was made by the individual investigator. Of the 292 implanted patients, 247 completed their 1-month follow-up visit, 230 completed their 3-month follow-up visit, and 214 patients completed their 6-month follow-up visit. There were a total of 11 deaths during the course of the study (seven with dedicated leads and four with integrated leads) that were reported to be unrelated to the device. There were no cases of RV lead dislodgement but there were two cases of RV lead perforation during the course of the study (one with dedicated lead and one with integrated lead). The remainder of the dropout was due to withdrawal from the study or noncompliance with follow-up. The drop-out rates between the two randomized groups were similar (26.5% with dedicated leads vs. 27.4% with integrated leads). Table I. Patient Population Dedicated Integrated Leads Leads (n = 147) (n = 145) Age (years) 69 ± ± 12 (Mean ± SD) Gender 72% male 76% male Ejection fraction (%) 25 ± 7 24 ± 7 (Mean ± SD) NYHA class (%) II 8 8 III IV 5 1 Unknown 3 3 Coronary artery disease (%) RV Lead Position (%) RV apex ICD Implant Indication (%) Primary Secondary Unknown 8 5 Anodal Stimulation The incidence of anodal stimulation at any LV output up to 7.5 V is shown in Figure 3. At implant, anodal stimulation was seen in 64% of patients with dedicated bipolar RV leads compared to only 1% with integrated bipolar RV leads (p < ). The incidence of anodal stimulation with the dedicated leads fell progressively to 51.5% at 6 months postimplant (p = 0.04), but always remained significantly higher than with the integrated leads. The incidence of clinically relevant anodal stimulation (anodal stimulation at a voltage less Figure 3. Prevalence of anodal stimulation, significantly lower with integrated bipolar RV lead systems as compared to dedicated bipolar RV lead systems. 160 February 2009 PACE,Vol.32

5 INTEGRATED VS. DEDICATED DEFIBRILLATION LEADS Figure 4. Prevalence of clinically relevant anodal stimulation (anodal stimulation at a voltage less than or equal to twice the LV capture threshold), significantly lower with integrated bipolar RV lead systems as compared to dedicated bipolar RV lead systems. than or equal to twice LV capture threshold) is shown in Figure 4. Again, at each follow-up, clinically relevant anodal stimulation was significantly more common with dedicated bipolar RV leads than with integrated bipolar RV leads (P < at implant; P = 0.02 at the 6-month follow-up). The incidence of clinically relevant anodal stimulation with dedicated bipolar RV leads was 22% at implant and fell progressively (P = 0.005) during follow-up to 10% at 6 months postimplant. By contrast, the incidence of clinically relevant anodal stimulation with integrated bipolar RV leads never exceeded 4% at any time of follow-up. There was no significant difference in LV pacing threshold during cross-chamber pacing between patients with dedicated versus inte- grated RV leads, but LV cross-chamber pacing impedances were lower in patients with integrated compared to those with dedicated RV leads (P < 0.05) (Table II). RV pacing thresholds and RV pacing impedances were lower with integrated compared to dedicated RV leads (Table II). Of the 75 patients with anodal stimulation at any LV output at 1-month postimplant, the LV output programmed at that time was available for analysis in 59 patients. A comparison of the anodal stimulation threshold, LV capture threshold, and actual programmed LV output in those patients is shown in Table III. The mean anodal stimulation was 3.89 ± 1.79 V, mean LV capture threshold was 1.58 ± 1.46 V, and mean programmed LV output was 3.50 ± 1.34 V. In 31 of the 59 patients (27 with dedicated RV leads and four with integrated RV leads), the programmed LV output exceeded the anodal stimulation threshold. In these patients, it is likely that there was actually anodal stimulation in the RV ongoing on an ambulatory basis. Ventricular Sensing There was no significant difference in RV sensing threshold between patients with dedicated versus integrated RV leads (Table II). There was no documentation of ventricular undersensing in any patient. Three episodes of oversensing due to electromagnetic interference (EMI) were documented; a single episode was documented in one patient with dedicated bipolar RV lead and two episodes documented in a second patient with an integrated bipolar RV lead. In all three episodes, the EMI was detected in both the atrial and ventricular channels (Fig. 5). No episodes of oversensing due to myopotentials Table II. Comparison of LV Capture Threshold, LV Pacing Impedance, RV Pacing Impedance, RV Pacing Threshold, and RV Sensing Thresshold between the Dedicated and Integrated Lead Systems at Different Follow-Up Points Implant 1 Month 3 Months 6 Months LV capture threshold (V at 0.5 ms) Dedicated 1.5 ± ± ± ± 1 Integrated 1.5 ± ± ± ± 1.4 LV pacing impedance (Ohms) Dedicated 713 ± 165* 690 ± 142* 705 ± 157* 696 ± 143* Integrated 609 ± ± ± ± 195 RV pacing impedance (Ohms) Dedicated 519 ± 91* 478 ± 68* 485 ± 75* 484 ± 96* Integrated 417 ± ± ± ± 84 RV pacing threshold (V at 0.5 ms) Dedicated 0.6 ± 0.2* 0.9 ± 0.8* 0.8 ± 0.6* 0.9 ± 0.7* Integrated 0.5 ± ± ± ± 0.4 RV sensing threshold (mv) Dedicated 10.8 ± ± ± ± 2.0 Integrated 10.3 ± ± ± ± 1.7 All values are mean ± standard deviation. * P < 0.05 compared to integrated leads. PACE,Vol.32 February

6 FREEDMAN, ET AL. Table III. Comparison of Anodal Stimulation Threshold, Left Ventricular (LV) Capture Threshold, and Programmed LV Output in 59 Patients with Anodal Stimulation at 1-Month Postimplant* Anodal LV Capture Anodal Stimulation Threshold LV Stimulation Patient Lead Threshold (V) at Programmed at Programmed Number Type (V) at 0.5 ms 0.5 ms Output (V) Output 2 Dedicated No 5 Dedicated No 7 Dedicated No 8 Dedicated Yes 10 Integrated Yes 13 Integrated Yes 18 Dedicated No 20 Dedicated Yes 22 Dedicated Yes 24 Dedicated Yes 25 Dedicated No 33 Dedicated Yes 37 Dedicated Yes 42 Dedicated No 63 Dedicated No 69 Dedicated No 74 Dedicated No 85 Dedicated Yes 86 Dedicated Yes 89 Dedicated No 95 Dedicated No 118 Dedicated Yes 119 Dedicated Yes 121 Dedicated Yes 122 Dedicated Yes 123 Dedicated No 126 Dedicated No 129 Dedicated No 132 Dedicated No 145 Dedicated No 153 Dedicated No 157 Dedicated Yes 162 Dedicated No 165 Dedicated No 166 Dedicated Yes 176 Dedicated No 181 Dedicated Yes 183 Dedicated Yes 185 Dedicated Yes 189 Dedicated Yes 197 Dedicated Yes 200 Dedicated No 213 Integrated Yes 216 Dedicated Yes 219 Integrated Yes 243 Dedicated Yes Continued. 162 February 2009 PACE,Vol.32

7 INTEGRATED VS. DEDICATED DEFIBRILLATION LEADS Table III. Continued Anodal LV Capture Anodal Stimulation Threshold LV Stimulation Patient Lead Threshold (V) at Programmed at Programmed Number Type (V) at 0.5 ms 0.5 ms Output (V) Output 248 Dedicated No 251 Dedicated No 252 Dedicated No 254 Dedicated Yes 270 Dedicated Yes 285 Dedicated No 301 Dedicated Yes 303 Dedicated Yes 322 Dedicated No 326 Dedicated No 327 Dedicated Yes 345 Dedicated No 371 Dedicated Yes Mean ± SD 3.89 ± ± ± 1.34 *An additional 16 patients had anodal stimulation at 1-month postimplant, but programmed LV output was not available for the analysis. were documented. There were no inappropriate ventricular tachyarrhythmia detections resulting from oversensing with either lead design. Far-Field R-Wave Sensing Twenty-five episodes of far-field sensing of R waves in the atrial channel were documented, occurring one or more times in 16 patients. Of these 16 patients, six had dedicated RV leads and 10 had integrated RV leads (P = NS). The far-field R-wave sensing invariably occurred during biventricular pacing and was detected as an inappropriate au- tomatic mode switch due to double counting of atrial rate. Discussion This prospective randomized trial compared integrated and dedicated bipolar RV defibrillator leads with respect to anodal stimulation and ventricular sensing performance. The principal results of this trial are (1) the incidence of anodal stimulation in integrated bipolar right ventricular leads was significantly lower than in dedicated bipolar right ventricular leads and (2) clinically Figure 5. Stored electrogram showing electromagnetic interference (EMI) of unknown origin detected as noise in a patient with a dedicated bipolar RV lead. Note that the EMI is detected in both the atrial and ventricular channels. Coincidentally, the EMI is detected shortly after the onset of ventricular tachycardia. RA = right atrial electrogram; RV = right ventricular electrogram. PACE,Vol.32 February

8 FREEDMAN, ET AL. detectable ventricular undersensing and oversensing were rare with both lead designs and did not differ between the two lead designs. The current study results are consistent with the findings of prior smaller studies. 11,12 In their study of 38 patients, Tamborero et al. found anodal stimulation during cross-chamber LV pacing in 74% of patients with dedicated bipolar RV leads versus none in patients with integrated bipolar RV leads. Thibault et al. found RV capture during cross-chamber LV stimulation in all 11 patients with CRT-D utilizing dedicated bipolar RV leads but none of 15 patients with integrated bipolar RV leads. Unlike the present study, neither Tamborero et al. nor Thibault et al. examined the likelihood of anodal stimulation at clinically relevant pacing outputs or the sensing characteristics of their leads. We found clinically relevant anodal stimulation in a significant proportion of patients with dedicated bipolar RV leads but only rarely in patients with integrated bipolar RV leads. We found no difference in RV sensing characteristics between the two lead designs. The differing incidence of anodal stimulation between dedicated and integrated bipolar RV leads is likely to be related to the differences in the surface area of the anodes. Due to the smaller surface area of an RV ring anode in a dedicated bipolar lead compared to the shocking coil anode in an integrated bipolar lead, current density in the area of the anode would be predicted to be higher with the dedicated bipolar lead than with the integrated bipolar lead. The higher current density would result in a greater likelihood of local myocardial depolarization at the anode with the dedicated bipolar leads. We found a progressive reduction in the incidence of anodal stimulation with cross-chamber LV pacing in dedicated leads following implantation. This trend is likely explained by the formation of a virtual electrode around the RV ring electrode with development of a conductive fibrotic capsule. The effective surface area of this virtual electrode is larger than the original electrode as it includes the original electrode and the newly formed fibrotic tissue. An increase in the virtual electrode size with time would be expected to progressively lower the current density around the anode and decrease the likelihood of anodal stimulation. RV pacing is often programmed to occur after a short delay following LV pacing in order to optimize the mechanical and hemodynamic benefit of CRT. If anodal RV stimulation occurs during cross-channel LV pacing, then this intended delay will not occur. In defibrillator systems incorporating a bipolar LV lead, anodal stimulation can be avoided by configuring LV pacing between the two LV electrodes rather than programming the system to cross-chamber LV pacing. This option is not available in patients where the LV lead is unipolar. Furthermore, even in patients with bipolar leads, cross-chamber LV pacing configuration is sometimes desired because of more favorable capture thresholds or better avoidance of phrenic nerve stimulation. Some newer resynchronizing defibrillator generators utilize the distal shocking coil as the anode for cross-chamber LV pacing even when the RV defibrillator lead is a dedicated bipolar design. In a study of 101 patients, Weretka et al. demonstrated that ventricular oversensing occurred more frequently in patients implanted with an integrated bipolar RV lead system (21/52) as compared to a dedicated bipolar RV lead system (4/49), with T-wave oversensing, P-wave oversensing, and diaphragmatic myopotential oversensing all demonstrated in patients with the integrated leads. 13 By contrast, in this study, ventricular oversensing was rarely seen, regardless of the lead design. Unlike the patients in this study, all the patients in the study by Weretka et al. underwent ambulatory monitoring, and this may partially explain the higher incidence of oversensing observed. However, there were inappropriate ventricular tachyarrhythmia therapies in 4% and repetitive inappropriate capacitor charges in 8% of the patients in that study, versus none in our study, and this difference cannot be attributed to difference in follow-up methodology. It is possible that the increased incidence of oversensing and inappropriate arrhythmia detections in the Weretka et al. study may be due to the difference in sense amplifier filters in the defibrillator generators used in that study (Boston Scientific, Inc., and Medtronic, Inc.) compared to the St. Jude Medical generators used in this study. Limitations Approximately 25% of the patients either withdrew from the study or were lost between implant and 6 months. It is conceivable that some bias was introduced into the study s results due to selective dropout of patients with a different likelihood of anodal stimulation or sensing performance. Only electrograms stored in the generator were available to examine the occurrence of ventricular undersensing or oversensing. Continuous ambulatory monitoring was not performed in these patients, and this might have detected a higher rate of sensing abnormalities. Clinical outcomes measured in this study were confined to measures of lead performance 164 February 2009 PACE,Vol.32

9 INTEGRATED VS. DEDICATED DEFIBRILLATION LEADS as well as oversensing, undersending, and inappropriate tachyarrhythmia detections, and followup in the study was terminated after 6 months. It is possible that longer follow-up would have revealed significant differences in clinical outcome in patients with one lead design versus the other. Also, the specific clinical impact of anodal stimulation and possible interference with LV-RV pacing offset were not examined in this study. The study utilized specific leads and pulse generators from a single manufacturer. Confining the study to defibrillator systems from a single manufacturer has the advantages of providing uniform lead designs, sensing amplifier filters, arrhythmia detection algorithms, and electrogram storage capability. However, the results cannot be necessarily extrapolated to defibrillator systems from other manufacturers. Conclusion We found that the incidence of RV anodal capture during LV cross-chamber pacing is significantly lower using an integrated lead system when compared to a dedicated lead system in patients receiving CRT-D systems. Oversensing and undersensing were uncommon with both dedicated and integrated RV lead systems. No patient had inappropriate ventricular arrhythmia detection or received inappropriate therapy due to oversensing of myopotentials or other electrical artifact. At the time of implantation, the physician s choice of a dedicated versus an integrated lead RV should take into account the clinical implications of the performance of the two lead designs. Following implantation, programming of LV pacing configuration and output should be done with the recognition of the possibility of anodal stimulation and its clinical implications. References 1. Salukhe TV, Francis DP, Sutton R. Comparison of medical therapy, pacing and defibrillation in heart failure (COMPANION) trial terminated early; combined biventricular pacemaker-defibrillators reduce all-cause mortality and hospitalization. Int J Cardiol 2003; 87: Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, Kocovic DZ, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346: Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352: Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R, Domanski M, et al. Amiodarone or an implantable cardioverter defibrillator for congestive heart failure. N Engl J Med 2005; 352: Hohnloser SH, Kuck KH, Dorian P, Roberts RS, Hampton JR, Hatala R, Fain E, et al. Prophylactic use of an implantable cardioverterdefibrillator after acute myocardial infarction. N Engl J Med 2004; 351: Van Gelder BM, Bracke FA, Meijer A. The effect of anodal stimulation on V-V timing at varying V-V intervals. Pacing Clin Electrophysiol 2005; 28: Mehra R, Furman S, Crump JF. Vulnerability of the mildly ischemic Appendix The following study centers and investigators participated in this trail: University of Utah Hospital and Clinics and Salt Lake VA Health Sciences Center, Salt Lake City, UT: Roger Freedman; East Texas Medical Center, Tyler, TX: Alex Petrakian; FirstHealth Moore Regional Hospital, Pinehurst, NC: Ker Boyce; St. Elizabeth s Medical Center of Boston, Boston, MA: Charles Haffajee; Galichia Heart Hospital, Wichita, KS: Jesus E. Val-Mejias; St. Luke s Methodist Hospital, Cedar Rapids, IA: Todd Langager; Lenox Hill Hospital, New York, NY: Noelle Langan; Fallon Clinic, Worcester, MA: Sanjeev Goyal; Montefiore Medical Center Moses ventricle to cathodal, anodal, and bipolar stimulation. Circ Res 1977; 41: Mehra R, McMullen M, Furman S. Time dependence of unipolar cathodal and anodal strength-interval curves. Pacing Clin Electrophysiol 1980; 3: Bennett JA, Roth BJ. Time dependence of anodal and cathodal refractory periods in cardiac tissue. Pacing Clin Electrophysiol 1999; 22: Bulava A, Ansalone G, Ricci R, Giannantoni P, Pignalberi C, Heinc P, Lukl J, et al. Triple-site pacing in patients with biventricular device Incidence of the phenomenon and cardiac resynchronization benefit. J Inter Card Electrophysiol 2004; 10: Tamborero D, Mont L, Alanis R, Berruezo A, Tolosana JM, Sitges M, Vidal B, et al. Anodal capture in cardiac resynchronization therapy implications for device programming. Pacing Clin Electrophysiol 2006; 29: Thibault B, Roy D, Guerra PG, Macle L, Dubuc M, Gagné P, Greiss I, et al. Anodal right ventricular capture during left ventricular stimulation in CRT-implantable cardioverter defibrillators. Pacing Clin Electrophysiol 2005; 28: Weretka S, Michaelsen J, Becker R, Karle CA, Voss F, Hilbel T, Osswald BR, et al. Ventricular oversensing: A study of 101 patients implanted with dual chamber defibrillators and two different lead systems. Pacing Clin Electrophysiol 2003; 26: Division, Bronx, NY: Jay Gross; Austin Heart, Austin, TX: David Kessler; St. Peters Hospital, Olympia, WA: Robert Wark; Hillcrest Hospital, Mayfield Heights, OH: Martin Wiseman; Roper Hospital, Charleston, SC: Brett Baker; Cardiology Associates of Marin and San Francisco, Larkspur, CA: Steven Hao; VA Medical Center Minneapolis, Minneapolis, MN: Venkat Tholakanahalli; Albany Medical Center, Albany, NY: James O Brien; St. Luke s Hospital Roosevelt, New York, NY: Jonathan Steinberg; Midwest Heart Foundation, Lombard, IL: Ray Kawasaki; Saint Barnabas Medical Center, Livingston, NJ: Marc Roelke. PACE,Vol.32 February