Predicting Asymptomatic Coronary Artery Diseasein Patients With Ischemic Stroke and Transient Ischemic Attack

Abstract

Background and Purpose— Identifying occult coronary artery stenosis may improve secondary prevention of stroke patients. The aim of this study was to derive and validate a simple score to predict severe occult coronary artery stenosis in stroke patients. Methods— We derived a score from a French hospital–based cohort of consecutive patients (n=300) who had an ischemic stroke or a transient ischemic attack and no previous history of coronary heart disease (Predicting Asymptomatic Coronary Artery Disease in Patients With Ischemic Stroke and Transient Ischemic Attack [PRECORIS] score) and validated the score in a similar Korean cohort (n=1602). In both cohorts, severe coronary artery stenosis was defined by the presence of at least 1 ≥50% coronary artery stenosis as detected by 64-section CT coronary angiography. Results— A 5-point score (Framingham Risk Score–predicted 10-year coronary heart disease risk [≥20%=3; 10–19%=1; <10%=0] and cervicocephalic artery stenosis [≥50%=2; <50%=1; none=0]) was predictive of occult ≥50% coronary artery stenosis risk in the derivation cohort (C-statistic=0.77 [0.70–0.84]) and in the validation cohort (C-statistic=0.66 [0.63–0.68]). The predictive ability of the score was even stronger when only ≥50% left main trunk disease or 3-vessel disease were considered (C-statistic=0.83 [0.74–0.92] and 0.70 [0.66–0.74] in derivation and validation cohorts, respectively). The prevalence of occult ≥50% coronary artery stenosis and ≥50% left main trunk or 3-vessel disease increased gradually with the PRECORIS score, reaching 44.2% and 13.5% in derivation cohort and 49.8% and 12.8% in validation cohort in patients with a PRECORIS score ≥4. Conclusions— The PRECORIS score can identify a population of stroke or transient ischemic attack patients with a high prevalence of occult severe coronary artery stenosis.