Keywords
INTRODUCTION
The use of implantable automatic defibrillators (ICD) with antitachycardia pacing (ATP) therapies has been demonstrated to reduce the number of discharges from the device.1-3 The efficacy of ATP ranges from 80% to 94%4-12 and can be affected by factors such as baseline cardiopathy, presence of necrosis due to previous infarctions, left ventricle (LV) size, and to some extent sympathetic tone prior to tachycardia.5,6,11,13
The appearance of ventricular resynchronization therapy and the opportunity for multiple site and cardiac chamber stimulation opens up a wide range of alternatives for ATP. We should first evaluate the influence of site of stimulation on ventricular tachycardia suppression by ATP. In patients with heart failure and substantial LV dilatation, single site ATP on the right side can be less effective than expected. Cardiac resynchronization therapy (CRT) enables us to apply these treatments from the LV wall or simultaneously from both ventricles. A previous study reports biventricular pacing proved more effective at suppressing ventricular arrhythmias than isolated apical right ventricular (RV) pacing.14
Cardiac resynchronization therapy can contribute to ATP efficacy through positive ventricular remodeling by reducing stress on the wall and facilitating ATP which, perhaps, reduces the recurrence of ventricular tachycardia (VT).15-17 However, the usefulness of CRT in reducing recurrent ventricular arrhythmias over time has not been definitively established18 and appropriate prospective studies should be designed to determine this.
The objective of this study is to investigate the influence of resynchronization therapy on ATP efficacy in suppressing VT episodes by comparing ATP efficacy in a group of patients with CRT versus another group without.
PATIENTS AND METHOD
Study Design
The VENTAK CHF/CONTAK CD study had a 2-phase design19-22 and both parts were used for the ATP arrhythmia suppression efficacy substudy. The original design appears in Figure 1. We enrolled patients presenting symptomatic heart failure (New York Heart Association [NYHA] class III-IV); indicated for ICD following guidelines of the American College of Cardiology; with ejection fraction (EF) <35%; QRS complex >120 ms; and optimized medical treatment for 1 year. We excluded patients indicated for pacemaker implantation for bradycardia. All participants gave written informed consent and the study was approved by the trial committees of each participating institution.
Figure 1. Initial VENTAK CHF/ CONTAK CD study design. Randomization is prior to implantation. After a 1-month recovery period, devices are randomly programmed to biventricular pacing (BiV) or VVI at 40 beats/minute to avoid stimulation. After 3 months follow-up, patients change mode of stimulation and follow-up continues for another 3 months.
Patients included in the first phase received an ICD (VENTAK CHF/CONTAK CD, Guidant Corporation) with conventional electrodes for the right atrium (RA) and RV. Left ventricular pacing was initially achieved through epicardial placement via left thoracotomy. This was later replaced by intravenous placement of an electrode with stimulation and sensing functions (EASYTRAK, Guidant Corporation) through the coronary sinus. At 30 days post-implantation, devices were programmed appropriately in each therapy group.
Half the patients were randomized to biventricular pacing in VDD mode and half to VVI at 40 beats/minute to avoid stimulation. At 3 months, therapies were reversed. After a second 3-month follow-up phase, further programming changes were made at the discretion of the researcher.
In 1999, the study design was modified to permit stabilization of the therapy groups (Figure 2). The cross-sectional design was replaced by 2 independent therapy groups: 1 with biventricular pacing and the other without stimulation for a period of 6 months.
Figure 2. Initial design modification. At the request of the Food and Drug Administration the initial cross-sectional design was modified and a parallel design was adopted with a 6 month follow-up. BiV indicates biventricular.
Antitachycardia Pacing Efficacy Substudy
We enrolled patients from both phases of the principal study. From the first phase we included VT episodes occurring during the first 3 month follow-up; for patients included in the second phase, we analyzed episodes occurring over the first 6 months (Figure 3). The objective of the substudy was to analyze ATP therapy efficacy with and without CRT.
Figure 3. Our study included patients during the first period of randomization, so half have a 3 month follow-up and the other half a 6 month follow-up. BiV indicates biventricular.
We enrolled 490 patients with heart failure and indication for ICD in the ATP substudy: half (245 patients) with CRT and half without. Baseline cardiomyopathies among these patients were ischemic heart disease (69%) and nonischemic dilated cardiomyopathy (31%). Mean EF was 22.3±7.9%. We found 33% were in NYHA functional class II, 58% in class III and 9% in class IV. We excluded NYHA class I patients.
We recorded previous monomorphic VT at implantation in 45% of patients, nonsustained VT in 26%, polymorphic VT in 13%, and ventricular fibrillation in 13%. Wide QRS complex was caused by left bundle branch block in 54% of patients and right bundle branch block in 13%. We found non-specific anomalies of intraventricular conduction in 33%.
Angiotensin enzyme conversion (ACE) inhibitors or angiotensin receptor blockers (ARA-II) were being administered to 87% of patients, diuretics 85%, digoxin 68%, and beta-blockers 47%.
From this population we selected 32 patients receiving ATP therapies (15 with CRT, 17 without CRT) because of VT episodes during the study period. To evaluate ATP efficacy, we analyzed all VT episodes stored in device memories of patients who received ATP therapy at least once.
Materials
Initially, LV stimulation was from an epicardial electrode, implanted via left thoracotomy, with LV lateral wall implantation as our first option. We first used the VENTAK CHF ICD, which permits simultaneous stimulation in both ventricles with the epicardial electrode and a standard defibrillation electrode (ENDOTAK, Guidant). Later, we deployed an endocardial electrode in the coronary sinus, implanted for stimulation and sensing in the LV. This electrode is passed through the coronary venous system over a guidewire. Left and right ventricular electrodes are connected in parallel so that stimulation and sensing are simultaneous. This parallel arrangement means ATP is administered from both electrodes. Consequently, ATP therapies were biventricular in all patients, independently of the final programming assigned.
Statistical Analysis
Continuous variables are expressed as mean±standard deviation (SD). The proportion of successful monomorphic VT conversions over time was analyzed with the Mantel-Haenszel test. Conversion rates at specific times in the 2 groups were compared with the Fisher exact test. Values of P≤.05 were considered statistically significant.
RESULTS
Of 490 patients, 32 received ATP therapies: 15 in the group with CRT and 17 in the group without. All ATP episodes were stored in the device memory permitting observation of intracavity electrograms to evaluate therapy efficacy. In the first month, global ATP therapy efficacy was 73%, rising to 93% in the second month, 86% in the third and 89% in patients with >3 month follow-up. These data show a significant increase in long-term ATP efficacy (P<.004) (Figure 4).
Figure 4. Global antitachycardia pacing efficacy. Efficacy is 73% in the first month and increases significantly during the study.
In patients with CRT, we recorded 221 episodes of ventricular arrhythmias treated with ATP, with a conversion rate of 90.5%. In the group without CRT we recorded 139 episodes of ATP, with only 69.1% efficacy (P<.001) (Table 1).
Some patients in both groups presented episodes of arrhythmic storms during follow-up. Consequently, many episodes were concentrated in a small number of patients which may constitute a bias when evaluating the results. To minimize this, in the final results we repeated the analysis excluding all patients presenting >30 episodes: 3 with CRT and 1 without CRT. The result was less striking but still statistically significant (P<.004) with a higher conversion rate among patients with CRT (Table 1).
We also evaluated ATP efficacy as a function of electrode site. We found 51% of leads were implanted in the lateral wall, 29% in the anterior wall, 14% in a posterior vein, and 5% in the apex. We found no electrode sites in the middle cardiac vein or coronary sinus. Table 2 shows ATP efficacy at 6 months in the CRT group as a function of electrode position. We found a trend towards higher ATP therapy conversion rates in lateral versus anterior wall sites. This probably reflects the greater hemodynamic benefit of lateral wall stimulation.23,24 Analysis of ATP efficacy in the lateral vein showed it was greater in the with CRT group when compared with lateral implantation in the without CRT group suggesting efficacy may be more closely related with the effect of ventricular remodeling than with the ATP administration site. However, due to the limited number of episodes recorded in both groups we did not find statistically significant differences.
We do not provide data on posterior wall sites due to the very limited number of patients in this group.
DISCUSSION
Our study shows ATP efficacy is greater in patients with CRT and that it tends to improve over time.
Moreover, the effectiveness of this biventricular treatment is comparable to that described in previous reports, when it is only applied from the RV apex.4-12
Previous studies have shown biventricular pacing can reduce the inducibility of ventricular arrhythmias,17 ventricular extrasystoles,16 or incidence of spontaneous ventricular arrhythmias15. Higgins et al15 analyzed the first patients included in the VENTAK CHF study and found episodes of tachycardia treated in 13 of 32 patients (41%). Five patients (16%) had 1 episode during biventricular pacing while 11 (34%) presented an event in the non-stimulation phase. Three patients (9%) received therapies in both phases. The difference between the number of patients needing therapy in the 2 groups was statistically significant (P<.0035). In a group of 20 patients, Walker et al16 showed that biventricular pacing reduced the number of ventricular extrasystoles over time by comparison with patients in sinus rhythm and RV apical stimulation. Zagrodzky et al17 showed a reduction in the inducibility of significant arrhythmias with biventricular pacing by comparison with RV pacing. However, the efficacy of biventricular pacing in the long-term reduction of arrhythmias has yet to be determined.
The InSync ICD study14 showed that ATP therapies in biventricular mode could be more effective than those administered from RV apex only. The study recorded 472 episodes of VT or fast ventricular tachycardia in 26 patients; 339 episodes in 17 patients in the VT zone, 107 episodes in 8 patients in the fast VT and 26 in 8 patients in the fast VT zone. Biventricular pacing with ATP was more effective in the VT and fast VT zones (P<.001).
Due to the device design employed, all ATP therapies in our study were applied simultaneously from both ventricles. Consequently, conversion rate efficacy in both groups can only be explained by the mid- and long-term effects of resynchronization on myocardial function. Various mechanisms may be involved: first, hemodynamic improvement in patients treated with CRT.25-27 Many studies have shown lower incidence of ventricular arrhythmias in patients with compensated heart failure.28-30 Moreover, biventricular pacing permits simultaneous activation of large areas of myocardium, permitting earlier, homogeneous repolarization. In these conditions ventricular extrasystoles are unlikely to find areas of functional block that facilitate reentry. Third, homogenization of refractory ventricular periods obtained via biventricular pacing. Finally, CRT would diminish sympathetic tone in patients with ventricular dysfunction31-33 and create a potent antiarrhythmic effect.33
Ideally, we would have long-term results although we can hypothesize that 3-6 month data are representative of long-term effects. Unpublished data from the Incidencia study confirm the effect of anatomic and electric remodeling in an early post-implantation phase. More interesting data would illustrate long-term ATP efficacy in patients without CRT due to worsening of their conditions over time. Data from this study are insufficient as the longest follow-up was only 6 months and in most patients randomized to the baseline group CRT was activated at the end of follow-up.
Antitachycardia Pacing Efficacy According to Electrode Site
Conversion efficacy is greater in lateral than anterior wall sites. Initially, 2 factors may contribute to this phenomenon: the importance of electrode site at the moment of biventricular ATP and remodeling due to LV biventricular pacing. We found that not only are lateral sites better than anterior sites in the CRT group but they are also better in the group without CRT. This indicates that the mechanism underlying efficacy is more closely related to ventricular remodeling than the final site of ATP administration. This is consistent with other studies23, 24 that show hemodynamic improvement is greater with LV stimulation from a lateral site by comparison with anterior stimulation.
CONCLUSIONS
Our results indicate that ATP therapy efficacy, when administered in both ventricles, is as high as in previous studies4-12,34 and that it improves during follow-up. There is a great difference between patients with cardiac resynchronization therapy and those without, with a trend that indicates efficacy is highly dependent on site of electrode stimulation and that the lateral site is the best option. All this probably reflects the effect of positive secondary myocardial remodeling on cardiac resynchronization.
Limitations
Our study presents some limitations. The first is the low number of episodes of VT treated with ATP during the follow-up. From 490 patients, we only selected 32 (6.5%). This has 2 explanations: the short follow-up period and the fact that device programming was optional. Unfortunately, many devices did not have ATP therapies activated. The final result is that the number of patients included is reduced, which somewhat undermines the soundness of the results.
Another limitation is the fact that we are dealing with a sub-group study, which means results should be analyzed with care.
See editorial on pages 1139-41
On behalf of the VENTAK CHF/CONTAK CD Study researchers and Guidant. The VENTAK CHF/CONTAK CD study received financial support from Guidant Corporation, St. Paul, Minnesota, USA.
Received January 3, 2005.
Accepted for publication June 15, 2005.
Correspondence: Dr. I. Fernández Lozano.
Unidad de Arritmias. Clínica Puerta de Hierro.
Avda. San Agustín de Porres, 4. 28029 Madrid. España.
E-mail: ifernandezl@sego.es