Observed Versus Predicted Stroke Risk After Transient Ischemic Attack: A Post Hoc Analysis of a Randomized Clinical Trial

Background The ABCD 2 score, used to triage patients with transient ischemic attack (TIA), was developed before contemporary use of dual antiplatelet therapy (DAPT) and therefore estimates pretreatment stroke risk. However, posttreatment stroke risk after DAPT may be more relevant for clinical triage. We aimed to quantify residual ischemic stroke risk following DAPT in patients with high‐risk TIA. Methods We conducted a post hoc analysis of high‐risk patients with TIA enrolled in the POINT (Platelet‐Oriented Inhibition in New TIA and Minor Ischemic Stroke) trial who were treated with DAPT. We calculated Kaplan–Meier estimates of ischemic stroke risk at 2, 7, and 90 days. Risk estimates were further stratified according to ABCD 2 category (4–5 versus 6–7) and presence of infarct on index imaging despite symptom resolution. Observed risks were compared with ABCD 2 predictions. Results A total of 1052 POINT participants experienced a TIA and were treated with DAPT. Cumulative ischemic stroke risks were 1.3% (95% CI, 0.6–1.9) at 2 days, 2.0% (95% CI, 1.2–2.9) at 7 days, and 4.1% (95% CI, 2.8–5.3) at 90 days, all below ABCD 2 ‐predicted risks of 5.1, 7.4, and 11.9%, respectively. Incident ischemic stroke risk did not differ by ABCD 2 category ( P =0.80) but was higher in those with acute infarct on index imaging ( P =0.02). In patients with infarct on imaging despite symptom resolution, 2‐day ischemic stroke risk (3.4% [95% CI, 0.1–6.1]) approximated ABCD 2 risk predictions, whereas those without evidence of infarct were at 6‐fold lower 2‐day risk (0.8% [95% CI, 0.0–1.4]). Conclusions High‐risk patients with TIA treated with DAPT had substantially lower early stroke risk than predicted by ABCD 2 , particularly in the absence of acute infarct on neuroimaging. Registration URL: https://www.clinicaltrials.gov ; Unique Identifier: NCT00991029.