The ICD: When Smarter is Better and Less is More: The VITALITY AVT INTRINSIC RV Study

Brian Olshansky, MD
Brian Olshansky, MD
Implantable cardioverter defibrillators (ICDs) have a clear-cut role in reducing the risk of sudden cardiac death in patients with underlying heart disease. Recent large randomized controlled clinical studies have demonstrated that ICDs prolong overall survival by preventing sudden cardiac death in patients with left ventricular dysfunction regardless of whether they have congestive heart failure, whether or not they had previously documented cardiac arrhythmias, or whether their heart disease was ischemic or non-ischemic in etiology. Seasoned cardiac electrophysiologists have known for quite some time that ICDs can benefit patients better than medical therapy alone. Those involved with device implantation also have noted a tremendous growth in the technological advancements in the ICD over the years. A recent trial highlighted the life-saving benefits of a relatively simple, single-chamber shock-box ICD, which has left questions of the relative benefits of today s more sophisticated technologically superior devices. In fact, some have suggested that the benefits of the simple shock box might even exceed those provided by more sophisticated devices. As such, CMS has indicated in their draft coverage decision that patients who undergo ICD implants for prophylactic purposes should have single-chamber devices unless there is justification for a dual chamber device (www.cms.hhs.gov/med) Why may this be the case? Let s start with why dual chamber devices are generally the preferred option for the electrophysiologist. Dual chamber devices have several important capabilities. For example, they can: 1) sense and pace in the atria, thereby providing physiologic pacing support; 2) sense and pace in the ventricles; 3) provide hemodynamically appropriate AV intervals; 4) help distinguish atrial from ventricular arrhythmias; 5) help treat and prevent atrial and ventricular arrhythmias; and 6) provide rate responsive pacing in the atria and the ventricles. This all appears to be good news, so what is the problem? Surely all of these additional capabilities argue against the simple shock-box being superior. Well, it may not be so simple. Several important clinical trials have demonstrated that dual chamber devices can be associated with poorer outcomes than single-chamber devices. One critical ICD study is the DAVID trial.1 In this trial, all patients received a dual chamber ICD, but the device was programmed to either a single chamber (VVI) with a rate floor of 40 beats per minute, or a DDDR with a rate of between 70-130 beats per minute. It is very important to note that in DAVID, these patients were selected specifically because they did not have a pacing indication. The patients programmed to VVI-40 experienced decreased heart failure hospitalizations and mortality (as a combined endpoint) compared to those programmed to DDDR, and the trial was stopped early because the difference was so great. Are these results surprising? Well, perhaps not when one realizes that in this study over 50% of the patients in the DDDR arm were paced unnecessarily in the right ventricle (RV) compared to less than 3% RV pacing in the VVI arm. Just as the resynchronization therapy trials have shown an improvement in heart failure morbidity and mortality, it is then not surprising that unnecessary RV pacing could result in desynchronization, which might be associated with the converse. Similar trends were noted in the MADIT II Trial.2 Although not randomized to different device types, those patients who received dual chamber ICDs had a trend toward a greater number of heart failure hospitalizations than those who did not. Those patients who received single-chamber ICDs had a greater rate of heart failure hospitalizations than those patients who did not receive ICDs at all. The latter point may be expected simply because patients lived longer with ICDs than those who did not receive ICDs (i.e., they live longer and therefore have a greater chance of experiencing heart failure due to the progressive nature of heart disease), but there is something more concerning than that. It was possible that those patients who received ICDs had a greater chance of developing heart failure perhaps because of unnecessary right ventricular pacing; several pacing trials support the concept that unnecessary right ventricular apical pacing is potentially bad for your health. The PASE3 and the MOST4 trials were randomized, multicenter controlled clinical trials, mainly of elderly patients, in which patients who had an indication for pacing had either a single-chamber or a dual chamber (DDD) pacemaker placed. While the outcomes were similar in each group, in patients with heart block, especially in the PASE trial, those patients with symptomatic bradycardia due to sinus node dysfunction fared better with a backup VVI device. It is likely that the reason for this was that with a dual chamber device, atrial sensing and tracking caused more right ventricular apical pacing and thus caused problems. In these trials, particularly the MOST trial, rates of atrial fibrillation (AF) were less with dual chamber pacing, but other outcomes were no better. It appears then, that the sophisticated dual chamber ICDs may have many benefits over a simple shock box, but that it is essential to avoid unnecessary right ventricular apical pacing that may have serious adverse consequences. In the patient with a narrow QRS complex (i.e., without a bundle branch block), right ventricular apical pacing creates an artificial left bundle branch block depolarization pattern, and this can lead to ventricular desynchronization, just the opposite of what occurs with cardiac resynchronization therapy, which uses both left and right ventricular pacing to activate both ventricles nearly simultaneously. In fact, data now suggest that unnecessary right ventricular apical pacing is associated with a worse prognosis than a left bundle branch block alone correcting for all other factors.5 Perhaps pacing from another site in the right ventricle, such as the outflow tract or the septum, would be associated with improved outcomes and less ventricular desynchronization, but this hypothesis has not been tested in a large number of patients and is not proven. What appears to be the case is that minimization of unnecessary right ventricular pacing would be the best prescription. So how can this be accomplished? Enter the advocates of the shock-box. One method would be to simply have a single-chamber device and minimize the pacing rate, but this is associated with a plethora of potential problems. In an ICD population, especially in patients who have congestive heart failure and require medications such as beta-blockers or in patients who have recurrent ventricular tachycardia requiring anti-arrhythmic medications, sinus bradycardia and/or AV block may occur, and AV block may occur post-MI. Thus, ventricular pacing may ultimately be required. It appears that what is needed is a device with the capability to provide atrial backup rate-responsive pacing (for beta-blocker induced bradycardia), to sense in the atria and pace the ventricle as needed (for rhythm discrimination and in the case of AV block), and to provide proper AV synchrony when needed (to improve cardiac output, and possibly reduce AF incidence). What is really required is a dual chamber ICD that gives the clinician flexible treatment options, including the option to reduce RV pacing. One way to utilize a dual chamber device is to have programming flexibility such that intrinsic AV conduction can occur under most circumstances unless RV pacing is needed. Of course, one could simply program a long fixed AV delay, but that methodology can interfere with tachycardia detection (which is the reason for the device implant in the first place) and does little to optimize ventricular filling and thus cardiac output. It would not make much sense to have an ICD device that only paces in the atrium without any ventricular backup pacing support for two reasons: 1) in order to sense life-threatening tachycardias effectively, a ventricular lead must be present (so why not use it to your advantage?); and 2) if AV block does occur, the patient has no device support available: sometimes RV pacing is necessary! A methodology that has been available and approved for several years to help minimize ventricular pacing, yet provide all the potential benefits of a sophisticated and flexible dual chamber ICD, is the concept of AV search hysteresis (AVSH). Enter the VITALITY AVT INTRINSIC RV study (Figure 1). The concept being tested in this study is directly related to the use of AVSH to minimize ventricular pacing. The hypothesis is that single-chamber VVI pacing, at a rate of 40, would be associated with the same outcome, with respect to heart failure hospitalization and mortality as a device programmed to DDDR with a rate of 60-130, as long as AVSH was programmed on with the right settings. We also expected in this trial that with DDDR AVSH, several important advantages would become apparent. In the VITALITY AVT INTRINSIC RV study, patients with a wide variety of clinical indications for ICDs (all presently accepted indications, which included patients with a pacing indication as well) could be enrolled in the study. During the first week of implantation, the device was programmed to DDDR AVSH rate of 60-130 beats per minute. If the percent of right ventricular pacing was less than 20 percent, then the patient would be randomized to VVI at 40 or DDDR 60-130 with AVSH turned on. We expected that over 80% of patients would have less than 20% of right ventricular pacing with the protocol required settings. With the initial protocol programming, approximately 67% of patients were pacing more than 20% of the time, but by adjusting the AVSH parameters, we observed that unnecessary RV pacing could be minimized appropriately with approximately 80% of all patients experiencing less than 20% pacing in the RV.6 The enrollment in the trial took place over 15 months. One hundred and fourteen sites participated in this trial, and enrollment included 1,566 patients. Enrollment is now complete. This study has the largest number of ICD patients compared to any other ICD trial to date. It was also the fastest enrolling ICD study ever conducted, which is probably an indicator of how seriously physicians view this issue, and also derives, in part, from the ease of conducting the trial. Participation was made simple via the use of the Vitality AVT ICD Family s My Patient Profiles that allows for preloading of defined programming changes and activation through one touch programming. In this instance, My Patient Profiles was used to provide the programming for the randomization arms to reduce the burden on clinicians and their staff and increase compliance. In addition, VITALITY AVT INTRINSIC RV is the largest device trial to date to utilize a web-based data collection system (MedNet Solutions Inc., Minneapolis). That meant the study was essentially paperless. Dynamic validation checking helped research coordinators in reducing simple errors. The large amount of data contained in the patient data disk from the programmer was uploaded and values automatically populated into the database. This resulted in a substantial time savings. The ease of use of this system contributed to a rapid enrollment phase while maintaining high data quality. Our special thanks to all investigators and their staff for their participation, but especially to Dr. Stephen Moore (Elyria Memorial, Ohio), Dr. Murray Rosenbaum (Cardiac Arrhythmia Services, Florida), and Dr. Lai Kok (VA Richmond, Virginia), our highest enrollers. Presently, we are in the follow-up phase. Patients will be seen at three, six, and 12 months for follow-up study with quality of life, ongoing clinical status, and upload of the data disk forming the majority of the data gathered, in addition to any hospitalizations or adverse events. Due to the large size of the trial, many secondary endpoints can be assessed, including the percent of atrial fibrillation (VITALITY AVT can record atrial fibrillation burden, even in the VVI mode), quality of life, the use of medications and follow-up, heart failure symptoms, complications of device implantation, and many other endpoints. One thing we do not know is the optimal programming to minimize ventricular pacing in different patient types. However, we will have some information regarding various low percentages of ventricular pacing to determine if there is any specific upper limit of the percentage of ventricular pacing that is associated with adverse consequences. We will also be able to determine if the percentage of ventricular pacing depends on the underlying medical condition being treated with the ICD. Results should be available for this trial in mid 2005. This will be the most definitive trial regarding testing the concept of minimization of ventricular pacing with a dual chamber ICD. Until final results are available, we do know that by using this technique of AV search hysteresis, we can effectively minimize ventricular pacing and still give patients access to state-of-the-art ICD technology. In the VITALITY AVT INTRINSIC RV study, we will have an observational group. This group consists of patients who had greater than 20% right ventricular pacing at the one-week visit. This group represents patients who likely have a concomitant pacing indication. Programming is left up to the investigator to optimize patients in this arm of the study. It will be interesting to see how these devices are programmed and what the outcomes are in this population. It is important to note that there are other programming possibilities to minimize unnecessary RV pacing, and chief among these are AAI pacing and DDI pacing. There are substantial limitations for both programming methodologies. With AAI pacing, there is no backup ventricular pacing, and it is likely in this population of patients that this would be required at least at times. With DDI pacing, there are limitations in AV synchrony, and the possibility of pacemaker syndrome certainly exists. Further, with DDI pacing, there is the possibility of under-detection of ventricular tachycardia. It is also worth noting that the precise settings of AV interval may have important hemodynamic consequences in select patients. The Vitality AVT INTRINSIC RV study is a seminal trial that will test the success of AVSH in allowing patients to have the advantages of the more sophisticated and flexible dual chamber ICDs while avoiding the consequences of unnecessary RV pacing. At this point, it appears that for patients who receive ICD therapy (without resynchronization therapy), minimizing unnecessary ventricular pacing is going to be the way to go in an ICD. This is the future of ICD therapy. This trial presently leads the way.