Derivation of the ALESSA Score for Post-Stroke Recurrent Ischemic and Bleeding Events

Study Questions:

Can a novel scoring system be derived and validated to predict early ischemic events and major bleeding after an acute ischemic stroke in patients with atrial fibrillation (AF)?

Methods:

The authors followed a cohort of 854 patients from the multicenter RAF study with acute ischemic stroke and AF between 2012 and 2014 in the derivation cohort. The outcome for this study assessed at 90 days included ischemic events (combination of stroke, transient ischemic attack, and systemic embolism) and hemorrhagic events (symptomatic intracranial bleeding and International Society on Thrombosis and Haemostasis [ISTH]-defined major external bleeding). The validation cohort included 994 ischemic stroke patients with AF seen between 2014 and 2016.

Results:

Older age (hazard ratio [HR], 1.06 for each year; 95% confidence interval [CI], 1.00-1.11) and severe atrial enlargement (HR, 2.05; 95% CI, 1.08-2.87) were predictors for early ischemic outcome events. Small lesions (≤1.5 cm) were inversely correlated with both major bleeding (HR, 0.39; p = 0.03) and ischemic events (HR, 0.55; 95% CI, 0.30-1.00). The ALESSA score consisted of 2 points for age ≥80 years, 1 point for age 70-79 years, 1 point for an ischemic lesion >1.5 cm, and 1 point for severe atrial enlargement. The C-statistic in the derivation cohort was 0.697 (95% CI, 0.632-0.763) for ischemic events and 0.585 (95% CI, 0.493-0.678) for major bleeding events. In the validation cohort, the C-statistic was 0.646 (95% CI, 0.529-0.763) for ischemic events and 0.407 (95% CI, 0.275-0.540) for bleeding events.

Conclusions:

The authors concluded that among acute stroke patients with AF, high ALESSA scores were associated with increased risk of ischemic events, but not major bleeding events.

Perspective:

Identifying subsequent ischemic and hemorrhagic risk for patients with AF who develop an ischemic stroke is critical to inform the decision to resume warfarin. The authors attempted to develop a risk score to predict early (90-day) events. Similar to many other AF risk scores, they had modest success predicting ischemic risk, but were less successful in predicting major bleeding risk. In general, most patients who suffer an ischemic stroke should be treated with long-term anticoagulation to prevent recurrent ischemic events. Newer-generation anticoagulants are associated with lower intracranial hemorrhage risk and may be a particularly good option for patients at risk for this most serious form of bleeding.


< Back to Listings