New ALERTS Data Support Use of Implantable Cardiac Monitor in High-Risk ACS Patients—With Caveats

WASHINGTON, DC (UPDATED)—An implantable cardiac monitor designed to alert high-risk patients that they are having an acute coronary syndrome event can not only reduce the rate of false positive situations but also identify a substantial number of silent MIs, according to additional data from the ALERTS trial.

The Guardian System (Angel Medical Systems; Eatontown, NJ) was approved by the US Food and Drug Administration last year for use in patients with prior ACS who remain at high risk for recurrent events, despite an advisory committee previously recommending against device approval based on a lack of efficacy data. The device resembles a pacemaker: it is implanted in the chest and is wirelessly connected with an external device that manages alerts as well as a system programmer that can retrieve and display data.

ALERTS, the pivotal US premarket approval study of the device that compared outcomes in 907 patients who received the device and had the alarm system either turned on or off for a period of 6 months, did not meet its primary efficacy endpoint and was halted early for incomplete and unreliable ECG data. But as lead investigator C. Michael Gibson, MD (Beth Israel Deaconess Medical Center, Boston, MA), explained to TCTMD, all patients were followed for “a few years” with the alarm function on, following the initial study period, which led to a “much more statistically powerful” data set.

Gibson, who presented these additional findings here today in a late-breaking trial session at CRT 2019, said, “The big take-home messages are that you had a lower false positive rate—that is when you have the alarm on you're less likely to go into the emergency room with symptoms alone that cause an unnecessary visit. And then the other message is that there were 42 MIs that were detected where it was just the alarm, there were no symptoms, and that's pretty powerful.”

Indeed, the false positive rate per patient-year was lower among patients who had devices with the alarm on versus off (0.164 vs 0.678; P < 0.001), and the positive predictive value with the alarm on was higher than with it off (25.8% vs 18.2%; P = 0.031). Additionally, the alarm system did trigger 42 patients with actual MI, but without any symptoms, to head to the emergency room.

“In a certain way, we’re trying to get people to the ER and abort an MI or stop the STEMI before that vessel really permanently closes down,” Gibson said. “The future could mean some devices that are less invasive, subcutaneous rather than implantable. Obviously, everything is moving toward a less and less invasive approach, and ultimately something wearable might be a possibility for these patients. Things are evolving in a rapid way.”

The data were simultaneously published in the Journal of the American College of Cardiology.

Cost-effectiveness, Safety Questions

The results from the extended ALERTS study were not surprising to Edward J. Schloss, MD (The Christ Hospital, Cincinnati, OH), who was not involved. “It seems quite reasonable that an implantable device can detect ECG changes associated with myocardial ischemia,” he told TCTMD in an email. “The real challenge is learning how to assess the significance of these changes, and then translate that into appropriate cost-effective treatment pathways that affect outcomes.”

At this point, Schloss said he would find it difficult to recommend the Guardian system in any patients given the lack of positive “meaningful” efficacy data. “I do think we have enough information to justify a larger outcomes-based trial, and if we were given the opportunity, I would be happy to participate,” he said. “Even if the device ultimately is proven to have positive effects on outcomes, we still need to address issues on cost and long-term safety.”

Gibson said he would currently recommend use of this device in high-risk patients, perhaps those with complex anatomy or a left main stent. “[Those] people are the kinds of patients who might benefit the most from it,” he noted.

In discussion following the presentation, panelist Lorenz Räber, MD, PhD (Bern University Hospital, Switzerland), agreed that preventing false positive hospitalizations is an attractive benefit of a device like the Guardian system, but asked: “How can we be sure that in the device group there were not NSTEMIs that were missed because they were wrongly assured that nothing would happen but instead they suffered from NSTEMI?”

“A lot of first-in-man work went into blowing balloons up and then looking at what happened with the device. When you blow a balloon up in the cath lab, it takes a while for the EKG to change. This is an intracardiac EKG, and within a couple heartbeats the ST segments deviate,” Gibson responded. He added that something that drives him “absolutely crazy” about the STEMI and NSTEMI classifications is that if a patient presents with a posterior MI with ST-segment depression in the anterior precordium, it’s deemed an NSTEMI even though the artery is occluded. “I’m much more in favor of talking about occluded arteries than the designation NSTEMI and STEMI. We really look for occlusive events here to get around some of that nomenclature,” he said.

On the other hand, session co-moderator Anthony N. DeMaria, MD (Sulpizio Cardiovascular Center at the University of California, San Diego), commented that false-positive device alerts are also a big issue, asking, “[What happens when] that the thing goes off and the patient gets scared to hell and it's an artifact?”

The FDA was also rightfully concerned about this issue initially, Gibson said. “The false-positive rate was 0.678 patients per patient-year [with the device off], so it’d be like two-thirds of a [single] patient coming in a year for false-positive symptoms. That number went down to 0.16, to about a quarter of it, with the device [on]. So rather than increase consumption, this actually gave you fewer false positives, so in conjunction with symptoms it could be more accurate.”

Without symptoms the false-positive rate was about 20%, he added, noting that this is balanced out by the silent MIs that were found by the device.

Looking forward, Schloss said he would “like to see more information about technical issues with the device. What is the lead—is it a commercially available pacer lead? What are the specifications of the sensor module device—size, battery life? What are the details of the algorithm? Is there programmability? Can the device be tuned to improve accuracy—ie, if a patient has false positives or false negatives, can we reprogram sensing algorithm to address this?”

Additionally, “there may be some potential to create overtesting and overtreatment of both true and falsely detected ischemic events as doctors react to alerts,” he added. “As a doctor that implants pacers and ICDs, I also know that many of the safety issues with these devices, such as device failure and infection, may not present until years after implant. If a patient with this device later develops indications for a cardiac rhythm device, the presence of an existing lead could affect the risk of future device implant. These concerns would need to be factored into the implant decision process.”

Lastly, Schloss said that while the device may not be cost-effective on its own, “I hope industry will consider incorporating this technology into existing pacer or ICD systems as an additional feature. This could be a natural fit and would address many of the safety and cost concerns of the standalone device.”

Notably, Angel Medical Systems filed for bankruptcy in January, but Gibson said he doesn’t think this will “have any material impact” on the future of the device.