Usefulness of Rhythm Monitoring Following Acute Ischemic Stroke

• Published in: The American journal of cardiology 2021-05, Vol.147, p.44-51
• Main Authors: Khurshid, Shaan, Li, Xinye, Ashburner, Jeffrey M., Lipsanopoulos, Ana T. Trisini, Lee, Priscilla R., Lin, Aimee K., Ko, Darae, Ellinor, Patrick T., Schwamm, Lee H., Benjamin, Emelia J., Atlas, Steven J., Singer, Daniel E., Anderson, Christopher D., Trinquart, Ludovic, Lubitz, Steven A.
• Format: Article
• Language: English
• Subjects: Aging; Antihypertensives; EKG; Electronic health records; Epidemiology; Fibrillation; Hazard assessment; Health hazards; Hospitalization; Hypertension; Ischemia; Monitoring; Mortality; Rhythm; Risk analysis; Risk assessment; Stroke; Telemedicine; Utilization
• ISSN: 0002-9149; 1879-1913
• DOI: 10.1016/j.amjcard.2021.01.038

•The rate of cardiac rhythm monitoring at 6 months after stroke is just over one-third.•Only 30% of patients with cryptogenic strokes receive implantable loop recorders.•Monitor utilization after stroke is not related to estimated atrial fibrillation risk. [Display omitted] We characterized monitor utilization in stroke survivors and assessed associations with underlying clinical atrial fibrillation (AF) risk. We retrospectively analyzed consecutive patients with acute ischemic stroke 10/2018-6/2019 without prevalent AF and assessed the 6-month incidence of monitor utilization (Holter/ECG, event/patch, implantable loop recorder [ILR]) using Fine-Gray models accounting for the competing risk of death. We assessed for predictors of monitor utilization using cause-specific hazards regression adjusted for the Cohorts for Heart and Aging Research in Genomic Epidemiology AF (CHARGE-AF) score, stroke subtype, and discharge disposition. Of 493 patients with acute ischemic stroke (age 65±16; 47% women), the 6-month incidence of monitor utilization was 36.5% (95% CI 31.7, 41.3), and 6-month mortality was 13.6% (10.4, 16.8). Monitoring was performed with Holter/event (n = 107; 72.3%), ILR (n = 34; 23.0%) or both (n = 7; 4.7%). Monitoring was more likely after cryptogenic (hazard ratio [HR] 4.53 [3.22, 6.39]; 6-month monitor incidence 70.6%) and cardioembolic (HR 2.43 [1.28, 4.62]; incidence 47.7%) stroke, versus other/undocumented (incidence 22.7%). Among patients with cryptogenic stroke, the 6-month incidence of ILR was 27.5% [18.5, 36.5]. Monitoring was more likely after discharge home (HR 1.80 [1.29, 2.52]; incidence 46.1%) versus facility (incidence 24.9%). Monitoring was not associated with CHARGE-AF score (HR 1.08 per 1-SD increase [0.91, 1.27]), even though CHARGE-AF was associated with incident AF (HR 1.56 [1.03, 2.35]). In conclusion, rhythm monitors are utilized after one-third of ischemic strokes. Monitoring is more frequent after cryptogenic strokes, though ILR use is low. Monitor utilization is not associated with AF risk.