When patients present with neurologic symptoms consistent with a transient ischemic attack (TIA), accurate prediction of future stroke risk is critical to deciding between hospital admission and discharge home. The ABCD2/ABCD2i scores are often used in an attempt to distinguish patients at high risk of stroke in the next 7 days from those at low risk; however, the poor predictive accuracy of these scores has been repeatedly demonstrated.

To address the need for a better tool, researchers developed the Canadian TIA Score in a prior study and set out to externally validate this score in a recent multicenter prospective cohort study. The score results range from -3 to 23, and are based on clinical presentation, laboratory values, electrocardiogram (ECG), and head computed tomography (CT) results.

Investigators enrolled 7607 adults (mean age 68 years) with TIA or minor stroke defined by discharge diagnosis. At the time of assessment, physicians calculated the Canadian TIA Score, ABCD2 score, and ABCD2i score. Patients were stratified as low (<1%), medium (1%-5%), and high (>5%) risk for stroke using each of the clinical decision tools. Primary outcomes were new stroke or carotid revascularization within 7 days of initial assessment. Most patients had hypertension and nearly 20% had diabetes.

Continue Reading

The Canadian TIA Score classified 16.3% of patients as low-risk and 11.6% as high-risk. Patients in the medium- and high-risk categories were older, more likely to have multiple comorbidities, and more likely to be taking antiplatelet agents at baseline.

A total of 182 primary outcomes occurred, with 108 patients diagnosed with subsequent stroke and 83 undergoing revascularization. Thirty-four patients were lost to follow-up at 7 days. The estimated event rate for the primary outcome in the low-risk group was 0.7% with an observed event rate of 0.5% (interval likelihood ratio 0.20; 95% CI, 0.09-0.44). The ABCD2 and ABCD2i scores classified no patients as low-risk for the primary outcomes.

In current practice, no clinical decision rule accurately predicts the risk of stroke in the ensuing 7 to 30 days after TIA symptoms. Importantly, in this study, the 34 patients lost to follow-up were not accounted for in the analysis; if we assume the worst-case scenario, this could significantly impact prognostic value. However, this external validation study of the Canadian TIA Score demonstrates this clinical decision rule has the potential to effectively stratify the risk of stroke or need for revascularization in the subsequent 7 days.

Depending on the clinician’s level of comfort and the local availability of resources, low-risk patients could be evaluated in the outpatient setting and high-risk patients could be managed with urgent neurology consultation in emergency departments. If nothing else, the data support abandoning the ABCD2 and ABCD2i scores.

Alan Ehrlich, MD, is a deputy editor for DynaMed, Ipswich, Massachusetts, and assistant clinical professor in family medicine, University of Massachusetts Medical School, Worcester.

DynaMed is a database that provides evidence-based information on more than 3000 clinical topics and is updated daily through systematic surveillance covering more than 500 journals.


Perry JJ, Sivilotti MLA, Emond M, et al. Prospective validation of Canadian TIA Score and comparison with ABCD2 and ABCD2i for subsequent stroke risk after transient ischaemic attack: multicentre prospective cohort study. BMJ. 2021;372:n49. doi:10.1136/bmj.n49