Atrial Fibrillation Screening With Implantable Loop Recorder Fails to Significantly Reduce Stroke Risk

Recently published post hoc findings from the LOOP randomized clinical trial (NCT02036450) showed that screening for atrial fibrillation (AF) using an implantable loop recorder (ILR) did not result in a significant decrease in ischemic or severe strokes compared with usual care. More than 1 in 3 strokes were classified as disabling or lethal, with a nonsignificant 31% reduction in this outcome by ILR screening.¹

Among a cohort of 6004 individuals randomly assigned to either standard of care (n = 4503) or ILR (n = 1501), 1027 participants were diagnosed with AF during follow-up (controls, 12% vs ILR, 32%), of whom 89% (n = 910) initiated anticoagulation, and 4.6% (n = 42) later discontinued the treatment. In total, 5.2% (n = 315) of the total cohort had stroke during follow-up, with a median modified Rankin Scale (mRS) score of 2 (IQR, 1-3) that was not different across the groups. The overall stroke rate was 1.02 per 100 person years (controls, 1.08 [95% CI, 0.95-1.23]; ILR, 0.86 [95% CI, 0.67-1.10]; hazard ratio [HR], 0.80 [95% CI, 0.61-1.05]; P = .10).

Previous observational studies have shown that cardioembolic or AF-related strokes have worse prognosis than strokes related to AF; therefore, investigators speculated that unscreened persons may be worse when having a stroke than those screened for AF and treated accordingly. Led by Søren Zöga Diederichsen, MD, PhD, cardiology fellow, Copenhagen University Hospital-Rigshospitalet, the analysis aimed to assess the severity and etiology of incident stroke and the importance of prior stroke in patients screened for AF vs usual care.

Phase 2 Results Highlight Milvexian's Potential to Reduce Stroke Risk

Milvexian at doses of 25 mg to 100 mg twice daily showed an approximate 30% relative risk reduction of ischemic stroke compared with placebo.

The study included individuals 70 years or older with no history of AF but diagnosed with hypertension, diabetes, heart failure, or previous stroke. Eligible participants were randomly assigned in a 1:3 ratio to receive a Reveal LINQ with remote monitoring and daily review of any arrhythmias (ILR group) vs usual care (control group). At baseline, both the control (17.6%) and ILR group (17.5%) had a similar number of patients with a history of prior stroke. Prior stroke, transient ischemic attack, or serious adverse event was found in 25.3% of controls and 24.7% of the ILR group.

A total of 123 participants (2.0%) had severe stroke according to mRS (controls, 2.2% vs ILR group, 1.5%), corresponding to 40% and 35% of all strokes in each group, respectively, with an overall event rate of 0.40 (95% CI, 0.33-0.47) per 100 person-years (controls, 0.64 [95% CI, 0.54-0.76]; ILR group, 0.56 [95% CI, 0.41-0.76]; HR, 0.87 [95% CI, 0.62-1.22]; P = .43). Mild stroke was reported in 192 participants (3.2%), with similar prevalence rates between the control group (3.3%) and those who received prior screening (2.9%).

In terms of stroke etiology, 272 participants (4.5%) had ischemic stroke (controls, 4.8% vs ILR, 3.7%), corresponding to 87% and 83% of all patients with stroke in the control and ILR groups, respectively, with an overall rate of 0.87 per 100 person years (95% CI, 0.78-1.00; P = .07). Cardioembolic strokes of undetermined source (ESUS) was found in 117 participants (2.0%), with an overall rate of 0.38 per 100 person-years (95% CI, 0.31-0.45). A total of 43 patients (0.7%) had hemorrhagic stroke, 45 if analyzed as first event opposed to the event with highest mRS score at recurrent stroke, with no difference between the groups.

In total, 16% (n = 49) of the participants admitted with stroke had been diagnosed with AF before their event, and a further 5 patients were diagnosed with AF on the same day as the event. The median mRS score was 3 (IQR, 1-6) vs 2 (IQR, 0-3) for patients with and without AF diagnosis, respectively (P <.001). Within the control group, the median mRS score was 4 (IQR, 2-6) vs 1 (IQR, 0-3) for patients with and without AF diagnosis, respectively (P <.001), whereas there was no statistical difference within the ILR group. Of the 49 patients with AF diagnosis before the stroke, 44 (90%) had already initiated anticoagulation (controls, 88% vs ILR, 94%), and these had a median mRS score of 2 (IQR, 1-5) compared with 5 (IQR, 2-6) in those with untreated AF (P = .50).