WASHINGTON –Although an implantable device for detecting myocardial infarction missed the primary composite outcome endpoint in a controlled trial, a newly completed extended analysis associated the device with a higher positive predictive value and a lower false positive rate when compared to sham control, according to data presented at CRT 2019, sponsored by MedStar Heart & Vascular Institute.
“Among high risk patients, this system may be beneficial in the identification of both symptomatic and asymptomatic coronary events,” reported C. Michael Gibson, MD, chief of clinical research in the cardiology division at Beth Israel Deaconess Hospital, Boston.
The implantable device (AngelMed Guardian System), which received Food and Drug Administration approval in April 2018, is designed to identify MI through detection of ST-segment elevations in the absence of an elevated heart rate. When the system detects an event during continuous monitoring, it sends a signal (internal vibration and auditory signal to an external monitor) designed to tell the patient to seek medical care.
The previously published multicenter and randomized ALERTS (AngelMed for Early Recognition and Treatment of STEMI) trial that tested this device was negative for primary composite endpoint of cardiac or unexplained death, new Q-wave MI, or presentation at the emergency department (ED) more than 2 hours after symptom onset (J Am Coll Cardiol. 2019 Feb 25. pii: S0735-109730237-2). In that trial 907 patients were fitted with the device and then randomized to having the device switched on or left off.
At 7 days, a primary endpoint was reached by 3.8% of those in the device-on group versus 4.9% of those in the device-off group, which was not significantly different.
Although the primary endpoint was not met, there were promising results. For example, in those who did have an occlusive event, patients in the device-on group had better preserved left ventricular function when evaluated after the event, a result consistent with earlier presentation in the ED and earlier treatment. In fact, 85% of patients with an MI in the device-on group presented to a hospital within 2 hours, compared with just 5% of those in the device-off control group during the initial study period.
More evidence of a potential clinical role for the device has now been generated in a new extended analysis. This analysis was made possible because patients in both of the randomized groups continue to wear the device, including those in the device-off group who had their devices activated after 6 months. There are now 3 more years of data of follow-up from those initially in the device-on group and those switched from the device-off group.
“So we started the clock over with a new statistical analysis plan and new endpoints,” Dr. Gibson explained. The FDA was consulted in selecting endpoints, particularly regarding evidence that the device did not increase false-positive ED visits.
There were numerous encouraging findings. One was that 42 silent MIs, which would otherwise have been missed, were detected over the extended follow-up. Another was that the annualized false-positive rate was lower in those with an activated device (0.164/year) when compared to the original device-off group (0.678/year; P less than .001). Lastly, the positive predictive value of an alarm during the extended follow-up was higher than that of symptoms alone among the original device-off group (25.8% vs. 18.2%).
The device was found safe. The rate of system-related complications was under 4%, which Dr. Gibson said is noninferior to that associated with pacemakers.
One of the potential explanations for the failure of the device to achieve the primary endpoint in the original trial was an unexpectedly low event rate, according to Dr. Gibson.
Even before this extended analysis, the FDA had accepted the potential benefits of this device as demonstrated in the approval last year. In the labeling, the device is called “a more accurate predictor of acute coronary syndrome events when compared to patient recognized symptoms alone and demonstrates a reduced rate over time of patient presentations without ACS events.”
“About 50% of patients wait more than 3 hours after the onset of symptoms before reaching an emergency room,” observed Dr. Gibson. Emphasizing the evidence that delay is an important predictor of adverse outcomes, he suggested the alarm device might be useful in accelerating care in some high risk groups.