Risk Stratification for Recurrence and Mortality in Embolic Stroke of Undetermined Source.

BACKGROUND AND PURPOSE: The risk of stroke recurrence in patients with Embolic Stroke of Undetermined Source (ESUS) is high, and the optimal antithrombotic strategy for secondary prevention is unclear. We investigated whether congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and stroke or transient ischemic attack (TIA; CHADS2) and CHA2DS2-VASc scores can stratify the long-term risk of ischemic stroke/TIA recurrence and death in ESUS. METHODS: We pooled data sets of 11 stroke registries from Europe and America. ESUS was defined according to the Cryptogenic Stroke/ESUS International Working Group. Cox regression analyses were performed to investigate if prestroke CHADS2 and congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or TIA, vascular disease, age 65-74 years, sex category (CHA2DS2-VASc) scores were independently associated with the risk of ischemic stroke/TIA recurrence or death. The Kaplan-Meier product limit method was used to estimate the cumulative probability of ischemic stroke/TIA recurrence and death in different strata of the CHADS2 and CHA2DS2-VASc scores. RESULTS: One hundred fifty-nine (5.6% per year) ischemic stroke/TIA recurrences and 148 (5.2% per year) deaths occurred in 1095 patients (median age, 68 years) followed-up for a median of 31 months. Compared with CHADS2 score 0, patients with CHADS2 score 1 and CHADS2 score >1 had higher risk of ischemic stroke/TIA recurrence (hazard ratio [HR], 2.38; 95% confidence interval [CI], 1.41-4.00 and HR, 2.72; 95% CI, 1.68-4.40, respectively) and death (HR, 3.58; 95% CI, 1.80-7.12, and HR, 5.45; 95% CI, 2.86-10.40, respectively). Compared with low-risk CHA2DS2-VASc score, patients with high-risk CHA2DS2-VASc score had higher risk of ischemic stroke/TIA recurrence (HR, 3.35; 95% CI, 1.94-5.80) and death (HR, 13.0; 95% CI, 4.7-35.4). CONCLUSIONS: The risk of recurrent ischemic stroke/TIA and death in ESUS is reliably stratified by CHADS2 and CHA2DS2-VASc scores. Compared with the low-risk group, patients in the high-risk CHA2DS2-VASc group have much higher risk of ischemic stroke recurrence/TIA and death, approximately 3-fold and 13-fold, respectively.

Age- and sex-specific analysis of patients with embolic stroke of undetermined source.

OBJECTIVE: To investigate whether the correlation of age and sex with the risk of recurrence and death seen in patients with previous ischemic stroke is also evident in patients with embolic stroke of undetermined source (ESUS). METHODS: We pooled datasets of 11 stroke registries from Europe and America. ESUS was defined according to the Cryptogenic Stroke/ESUS International Working Group. We performed Cox regression and Kaplan-Meier product limit analyses to investigate whether age (<60, 60-80, >80 years) and sex were independently associated with the risk for ischemic stroke/TIA recurrence or death. RESULTS: Ischemic stroke/TIA recurrences and deaths per 100 patient-years were 2.46 and 1.01 in patients <60 years old, 5.76 and 5.23 in patients 60 to 80 years old, 7.88 and 11.58 in those >80 years old, 3.53 and 3.48 in women, and 4.49 and 3.98 in men, respectively. Female sex was not associated with increased risk for recurrent ischemic stroke/TIA (hazard ratio [HR] 1.15, 95% confidence interval [CI] 0.84-1.58) or death (HR 1.35, 95% CI 0.97-1.86). Compared with the group <60 years old, the 60- to 80- and >80-year groups had higher 10-year cumulative probability of recurrent ischemic stroke/TIA (14.0%, 47.9%, and 37.0%, respectively, p < 0.001) and death (6.4%, 40.6%, and 100%, respectively, p < 0.001) and higher risk for recurrent ischemic stroke/TIA (HR 1.90, 95% CI 1.21-2.98 and HR 2.71, 95% CI 1.57-4.70, respectively) and death (HR 4.43, 95% CI 2.32-8.44 and HR 8.01, 95% CI 3.98-16.10, respectively). CONCLUSIONS: Age, but not sex, is a strong predictor of stroke recurrence and death in ESUS. The risk is ≈3- and 8-fold higher in patients >80 years compared with those <60 years of age, respectively. The age distribution in the ongoing ESUS trials may potentially influence their power to detect a significant treatment association.