Comparison of Risk Scores for the Prediction of the Overall Cardiovascular Risk in Patients with Ischemic Stroke: The Athens Stroke Registry

Abstract

BackgroundStratification of overall vascular risk in patients with ischemic stroke is important as it may guide management decisions. Currently available schemes have only modest prognostic accuracy. The TRA2°P score aids in vascular risk stratification in patients with previous myocardial infarction (MI). AimWe investigated whether the prognostic performance of TRA2°P can be extended in patients with ischemic stroke and whether it can improve the risk stratification made by CHA2DS2VASc and Essen-Stroke-Risk-Score (ESRS). MethodsWe analyzed the Athens Stroke Registry using Kaplan-Meier survival and Cox-regression analyses to assess if TRA2°P (in different categorizations) predicts the composite endpoint of stroke recurrence, MI or cardiovascular death. We compared its incremental predictive value over CHA2DS2-VASc and ESRS and calculated continuous net reclassification indices (cNRI). ResultsIn 2833 patients (followed for 9278 patient-years) and 776 events, there was decreased survival probability for TRA2°P-based high-risk patients compared to low-risk (log-rank-test P < .001), but the discriminatory power for the occurrence of the composite endpoint was only modest (Harrell's-C:.566, 95% CI:.545-.587). Combined with ESRS, TRA2°P conferred incremental discrimination (Harrell's-C:.544, 95% CI:.513-.574 versus .574, 95% CI:.543-.605 respectively, P = .049) and reclassification value (cNRI = 9.8%, P = .02). Combined with CHA2DS2-VASc, TRA2°P did not improve discrimination (Harell's-C:.578, 95% CI: .547-.608 versus .585, 95% CI:.554-.616, P = .738). ConclusionThe currently available prognostic scores have generally low performance to predict the overall cardiovascular risk in ischemic stroke patients. Further research is needed to improve vascular risk stratification in ischemic stroke patients.