Incidence, characteristics and outcomes in patients with embolic stroke of undetermined source: A population-based study

Abstract

Embolic stroke of undetermined source (ESUS) represents a subgroup of cryptogenic ischemic stroke (CS) distinguished by high probability of an underlying embolic mechanism. There are scarce population-based data regarding the incidence, characteristics and outcomes of ESUS. Consecutive patients included with first-ever ischemic stroke of undetermined cause in the previously published population-based Evros Stroke Registry were further subdivided into ESUS and non-ESUS CS. Crude and adjusted [according to the European Standard Population (ESP), WHO and Segi population] incidence rates (IR) for ESUS and non-ESUS CS were calculated. Baseline characteristics, admission stroke severity (assessed using NIHSS-score), stroke recurrence and functional outcomes [determined by modified Rankin Scale (mRS) scores], were recorded during the 1-year follow-up period. We identified 21 and 242 cases with ESUS (8% of CS) and non-ESUS CS. The crude and ESP-adjusted IR for ESUS were 17.5 (95%CI: 10–25) and 16.6 (95%CI: 10–24) per 100,000 person-years. Patients with ESUS were younger (p < .001) and had lower median admission NIHSS-scores (p < .001). Functional outcomes were more favorable in ESUS at 28, 90 and 365 days. ESUS was independently (p = .033) associated with lower admission NIHSS-scores (unstandardized linear regression coefficient: -13.34;95%CI: -23.34, −3.35) on multiple linear regression models. ESUS was not related to 1-year stroke recurrence, mortality and functional improvement on multivariable analyses. In conclusion we found that ESUS cases represented 8% of CS patients in this population-based study. Despite the fact that ESUS was independently related to lower admission stroke severity, there was no association of ESUS with long-term outcomes.