The ABCD2 score classified 45% as low-risk but 17.7% of them had acute stroke, showing poor performance to rule out stroke in ED observation patients.
Does the ABCD2 score safely rule out acute stroke in patients presenting with suspected TIA in an emergency department observation unit?
The ABCD2 score misclassifies nearly 20% of low-risk EDOU patients with suspected TIA who actually have an acute stroke, making it inappropriate for ruling out stroke in this setting.
Background: The ABCD 2 score is a validated risk stratification tool used to quantify the risk of stroke within 48 hours among patients presenting with suspected transient ischemic attack (TIA). Limited data exist regarding its performance among patients in an emergency department observation unit (EDOU). The objective of this study was to determine the safety and efficacy of the ABCD 2 score in a US EDOU cohort. Methods: We conducted an observational cohort study using the Wake Forest Emergency Medicine TIA Registry. Patients ≥18 years old who were evaluated for possible TIA in the EDOU of Atrium Health Wake Forest Baptist from July 1, 2021 to December 1, 2023 were accrued. This analysis included patients without a persistent neurologic deficit, except for amaurosis fugax, vertigo, or paresthesias, as determined by the consulting neurologist and attending emergency physician. Patients were risk-stratified using the ABCD 2 score as low- (0–3), moderate- (4–5), and high- (≥6) risk for short-term stroke. The primary safety outcome was index encounter stroke, as determined by the consulting neurologist and imaging. Efficacy was defined as the proportion of patients classified to the low-risk group. Negative predictive values and positive predictive values, and negative likelihood ratios and positive likelihood ratios for the ABCD 2 score were calculated for index stroke and reported with 95% confidence intervals (CI). Results: Of the 340 EDOU TIA patients, 55.9% (190/340) were female, 27.9% (95/340) were nonwhite, and the mean age was 67.1 ± 13.6 years. During the index visit, 21.8% (74/340) were diagnosed with acute stroke. The ABCD 2 score classified 45.0% (153/340) as low-, 45.6% (155/340) as moderate-, and 9.4% (32/340) as high-risk. Of those identified as low-risk, 17.7% (27/153) had an index acute stroke. Negative predictive values and negative likelihood ratios of the ABCD 2 score for index stroke were 63.5% (95% CI, 53.3–72.6) and 0.7 (95% CI, 0.5–1.1). Among patients classified as high-risk, 34.4% (11/32) had an index stroke, corresponding to a positive predictive value of 34.4% (95% CI, 20.9–50.9) and a positive likelihood ratio of 1.9 (95% CI, 1.0–3.7). Conclusions: Although the ABCD 2 score classified approximately half of patients as low-risk, nearly 20% of this group were diagnosed with an acute stroke during EDOU evaluation. These findings suggest that the ABCD 2 score is not appropriate for ruling-out acute stroke among EDOU patients as it would lead to a high rate of missed strokes.
Lumia et al. (2026) studied this question. The ABCD2 score classified 45% as low-risk but 17.7% of them had acute stroke, showing poor performance to rule out stroke in ED observation patients.