Measuring the Effectiveness of Wearable Defibrillators and Implantable Devices: EP Lab Digest Speaks with Jeffrey Olgin, MD

Measuring the Effectiveness of Wearable Defibrillators and Implantable Devices: EP Lab Digest Speaks with Jeffrey Olgin, MD About the VEST/PREDICTS Study Tell us about the new VEST/PREDICTS study. What are the main objectives of both parts of the study? The Vest Prevention of Early Sudden Death Trial/Prediction of ICD Therapies Study (VEST/PREDICTS) is an NIH-funded, multi-center, randomized clinical trial and an observational study wrapped into one — it is one study that answers two questions by combining two studies in a single cohort. The study also receives funding from Medtronic, Zoll-Lifecor and GE. VEST is a randomized trial of patients admitted to the hospital with a myocardial infarction (MI) and an ejection fraction (EF) of ≤35% measured >24 hours after the MI. Upon discharge from the hospital, they are randomized to receive a LifeVest^® wearable defibrillator (ZOLL Lifecor Corporation) and optimal medical therapy or optimal medical therapy alone. The participants are followed for a primary endpoint of total 2-month mortality. The main objective of VEST is to determine whether a wearable defibrillator vest can reduce mortality in the first 2 months after an MI, prior to when an ICD is indicated. Recent data from VALIANT, EPHESUS and DINAMIT studies show that these patients have between a 2.0-3.4% risk of sudden death in the first 2 months after their MI. DINAMIT demonstrated that early ICD implantation did not affect mortality. It is unclear whether the “invasive” nature of ICD implantation and DFT testing associated with the implant adversely affects remodeling and subsequent mortality or whether there is a high mortality from other causes that offset reduction from sudden death mortality. In either case, it is important to determine whether a non-invasive, less expensive means of preventing sudden cardiac death (SCD) is effective in this early period. A recent study of AEDs in the post-MI period did not show mortality benefit (HAT study) in the group that received the AED. PREDICTS is an observational study that follows on after VEST follow-up is complete. At 2 months (the conclusion of VEST), all patients will move into the PREDICTS phase of the study. In this phase, an echocardiogram will be performed in all participants. If their EF is still ≤35%, they will receive a Medtronic ICD. If their EF >35, they will receive a Medtronic Reveal DX or XT. Both the ICD and Reveal DX/XT have identical programming for detection of ventricular fibrillation (VF) and work with the Medtronic CareLink^® home monitoring system. The participants will receive a battery of risk stratification testing at baseline including: 1) Holter for heart rate variability, heart rate turbulence, deceleration capacity and a variety of other parameters; 2) Echo for systolic and diastolic function; 3) Stress test for ischemia and T-wave alternans; 4) Baroreflex sensitivity testing; 5) ECG; 6) Signal-averaged ECG; 7) DNA analyses; and 8) Serum analyses. Each of these will be repeated yearly for 3 years. Participants will be followed for a mean of 5 years for a primary outcome of 30 beats of VF/VT at rates ≥182 bpm. At the end of the study, we plan to develop a multi-modality algorithm taking into account clinical and testing results to determine 1- and 5-year risk of VT/VF resulting in an appropriate shock (i.e., fast VT/VF of at least 30 beats in duration). The idea will be similar to what is used for CAD using the Framingham Risk Score; with this algorithm, we anticipate being able to enter a variety of information into a computer algorithm (on a handheld device or website) and testing values and being able to tell the clinician what the likelihood of an appropriate shock is for a given patient. We will also look at a variety of other secondary endpoints such as inappropriate shocks, morbidity, other cardiovascular endpoints, total mortality, atrial fibrillation and many others. The rationale for PREDICTS is that somewhere between 70-80% of patients who receive an ICD for primary prevention according to MADIT II or SCD-HeFT criteria do not receive an appropriate shock from the ICD (i.e., they do not need the ICD during that time). In reality, it is likely less than that in the real world application of these studies. Clearly, EF is a very non-specific marker of ICD need. With this study, we hope to develop a better method for identifying patients at the highest and lowest risk of needing an ICD. In addition, we hope to develop tools to identify patients with EF >35% that are at high risk. How many are expected to be enrolled? Who is eligible to enroll? We anticipate enrolling 4,500 participants from at least 60 sites. Patients who are admitted to the hospital with an MI and an EF ≤35% at least 24 hours after the MI are eligible. They can have a pre-existing low EF prior to the MI. Obviously, patients with ICDs at the time of enrollment are not eligible. What is unique about the VEST/PREDICTS study? There are several unique aspects of the study. First, it is one of the first major studies that uses remote home monitoring to ascertain its endpoint. Secondly, it is a unique collaboration between the NIH and 3 additional industry partners. In addition to it being an NIH-funded study, ZOLL Lifecor, Medtronic and GE are all partnering with us to make this happen. A study of this magnitude and importance could not be done by the NIH alone or any of the industry partners by themselves. Thirdly, this is the first major, prospective risk stratification study that looks at a wide array of risk stratification testing, does so repeatedly (risk is certainly not static) and identifies patients at a relatively uniform point in time from their MI. Lastly, this will be the largest cohort of patients with continuous arrhythmia monitoring. How might this study change the way we detect ventricular arrhythmias? This trial will enable us to sit down with a patient who has had an MI and give them a pretty reasonable estimate of their likelihood of having an appropriate ICD shock if one is implanted. We will be able to categorize primary prevention patients into a high and low risk group to better identify those who most need an ICD. In addition, some of these algorithms and tests may be incorporated into implanted devices that can alert a patient or physician when they “cross the line” into a higher risk group. Why is there a need for a new trial on wearable defibrillators? There has never been a randomized trial of the wearable defibrillator to know whether it will reduce overall mortality in this patient population. While it may be reasonable to assume that it would reduce mortality, it is not clear from existing data that this percentage of sudden cardiac death is primarily due to ventricular fibrillation and whether that ventricular fibrillation is primary. In other words, we don’t know that defibrillation in this period will reduce mortality since there may be other underlying causes of apparent sudden death in this group (acute ischemia, pump failure, asystole, etc.). In addition, since the LifeVest is effective only if worn, we don’t know how compliance will play into the effectiveness. Other factors such as anxiety, discomfort, or alarms from the device may also play into the effectiveness. The medical field is littered with “former” therapies that seemed at the time to be correct, but when studied in a rigorous fashion taught us otherwise (e.g., CAST study, DINAMIT, AFFIRM). On the other hand, the LifeVest is currently not prescribed commonly for this indication. If VEST is positive, it could change practice to make the LifeVest a standard of care, and this would then become evidence-based practice. A positive study could also give physicians enough evidence to convince patients to wear something that is somewhat inconvenient (the LifeVest) for 2 months if the patients knew it would save their life. A study will make this more than theoretical and actually provide the evidence for such assertions. How soon might we see a tool that predicts which post-MI patients are at greatest risk of SCA? Also, could the results of this trial possibly reduce the number of people who receive ICDs every year? We are just starting to enroll patients now. We anticipate VEST finishing in about 3.5 to 4 years, and PREDICTS would finish about 3 years after that. Right now only a small fraction of those currently indicated for an ICD for primary prevention based on recent studies are actually getting them. It is anticipated that if our study can provide a more accurate way to determine who actually needs them, more of the right people will get an ICD. The overall number of ICDs is not what is important, but rather that the right people at the highest risk receive them. This will make patients more willing to have them implanted and make referring physicians more willing to refer patients. Right now the number of patients that need to be treated is a significant barrier for referral for many primary care cardiologists and internists.