Triaging TIA/minor stroke patients using the ABCD2 score does not predict those with significant carotid disease

Background: 'Rapid Access' TIA Clinics use the ABCD(2) score to triage patients as it is not possible to see everyone with a suspected TIA <24 h. Those scoring 0-3 are seen within seven days, while patients scoring 4-7 are seen as soon as possible (preferably <24 h). It was hypothesized that patients scoring 4-7 would have a higher yield of significant carotid disease.

Methods: Prospective study of correlation between Family Doctor (FD) or Emergency Department (ED) ABCD(2) score and specialist consultant Stroke Physician measured ABCD(2) score and prevalence of ≥50% ipsilateral carotid stenosis or occlusion in patients presenting with 'any territory' TIA/minor stroke or 'carotid territory' TIA/minor stroke.

Results: Between 1.10.2008 and 31.04.2011, 2452 patients were referred to the Leicester Rapid Access TIA Service. After Stroke Physician review, 1273 (52%) were thought to have suffered a minor stroke/TIA. Of these, both FD/ED referrer and Specialist Stroke Consultant ABCD(2) scores and carotid Duplex ultrasound studies were available for 843 (66%). The yield for identifying a ≥50% stenosis or carotid occlusion was 109/843 (12.9%) in patients with 'any territory' TIA/minor stroke and 101/740 (13.6%) in those with a clinical diagnosis of 'carotid territory' TIA/minor stroke. There was no association between ABCD(2) score and the likelihood of encountering significant carotid disease and analyses of the area under the receiver operating characteristic curve (AUC) for FD/ED referrer and stroke specialist ABCD(2) scores showed no prediction of carotid stenosis (FD/ED: AUC 0.50 (95%CI 0.44-0.55, p = 0.9), Specialist: AUC 0.51 (95%CI 0.45-0.57, p = 0.78).

Conclusions: The ABCD(2) score was unable to identify TIA/minor stroke patients with a higher prevalence of clinically important ipsilateral carotid disease.