Primary results from the DEFINE AFib clinical study have found that Medtronic’s LINQ family of insertable cardiac monitors (ICMs), paired with a novel algorithm, were able to detect atrial fibrillation (AF) episodes and properly risk stratify patients as high risk prior to an AF-related healthcare utilisation 80% of the time. Using artificial intelligence (AI)-based algorithms, the Reveal LINQ and LINQ II ICMs quantified AF burden—a measure of time a person spends in AF during a monitored period—to inform treatment decisions and help anticipate future healthcare needs.
These results were presented at the ongoing 30th annual AF symposium (16–18 January 2025, Boston, USA).
The DEFINE AFib clinical study enrolled 973 patients using an app-based enrolment feature, and characterised the impact of AF burden on patient outcomes and quality of life. Using the data, researchers built an algorithm capable of predicting patients’ risk of needing AF-related healthcare in the next 30-day period, in addition to predicting clinically meaningful reductions in patient-reported quality of life.
Results from the study showed that 22% of study participants who crossed into the high-risk threshold for the first time experienced an AF-related healthcare utilisation (mean time, 164±145 days) compared to 9% of patients in the low-risk group. According to Medtronic, these data support the conclusion that the AI-based analytics from the ICMs provide valuable information, particularly for those at a higher risk of an AF-related hospitalisation, clinic visit, or therapeutic intervention.
“The first-of-its-kind DEFINE AFib study leveraged a unique design that engaged patients from the very beginning,” said Jonathan Piccini (Duke University Hospital/Duke Clinical Research Institute, Durham, USA), chair of the DEFINE AFib clinical study steering committee. “We know that how much AF a patient experiences matters, but we don’t know how different durations or patterns impact the risk of future health events. Combining continuous rhythm monitoring with traditional risk factors has helped clarify how AF burden and patterns can inform risk, prioritisation and treatment decisions. Using upgraded AI-based algorithms and ICM data, physicians are better equipped to understand variance in patients’ AF patterns, offering the opportunity to provide the right patient with the right therapy at the right time.”
A recent press release from Medtronic also relays findings from a sub-analysis of the DEFINE AFib clinical study—presented at the 2024 European Society of Cardiology (ESC) congress (30 August–2 September, London, UK)—which showed “important differences” in performance between the LINQ family of ICMs and the Apple Watch for AF episode detection. Notably, 40% of AF episodes (n=191) occurred while the Apple Watch was not being worn; AF episodes can potentially occur at night while wearables are often taken off to recharge. In addition, when worn, the Apple Watch was only able to detect 26% of AF episodes lasting 75 minutes or more that the LINQ ICM detected.
“Wearables allow patients to capture more real-time heart health data than ever before, but medical grade technology—like the LINQ family of ICMs—is necessary to provide clinicians with an accurate and reliable way to detect and manage cardiac conditions like AF,” said Alan Cheng, chief medical officer of the Cardiac Rhythm Management business, which is part of the Cardiovascular Portfolio at Medtronic. “These findings also indicate that, while consumer-grade devices such as smartwatches and monitors can provide some insights into overall heart health, they are limited in their ability to screen for and help manage chronic conditions like AF. Medical-grade devices with continuous monitoring capabilities like ICMs are more appropriate.”
The DEFINE AFib clinical study used AI and machine-learning techniques to analyse changes in AF burden over time, and the ICM-based model separated individuals into high- versus low-risk AF-related healthcare utilisation groups. Medtronic notes that AF-related healthcare utilisation measures included clinical actions like ablation, cardioversion, initiation/intensification of rate or rhythm control medication, or progression to a pacemaker or implantable cardioverter-defibrillator (ICD).
