Incidence of transient ischemic attack and early stroke risk: validation of the ABCD2 score in an Italian population-based study.

BACKGROUND AND PURPOSE: The importance of transient ischemic attack (TIA) lies on the short-term risk of stroke, and the ABCD2 score may improve early stroke risk prediction. However, population-based studies are still needed. We aimed to provide data on TIA incidence and to evaluate the ABCD2 predictive ability for early recurrent stroke in a population-based study. METHODS: This study is part of a 2-year prospective community-based registry of all cerebrovascular events in the district of Udine (153 312 inhabitants), Friuli Venezia Giulia region, northeast of Italy, between April 1, 2007 and March 31, 2009. Multiple overlapping sources for finding cases were used, combining hot and cold pursuit. RESULTS: We identified 178 TIA, 161 (90.4%) of which were incident. The crude overall annual TIA incidence rate per 1000 residents was 0.52 (95% confidence interval [CI], 0.45-0.61). Incidence rate was 0.45 (95% CI, 0.31-0.65) when standardized to the 2007 Italian population and 0.25 (95% CI, 0.16-0.39) when standardized to the European standard population. Estimates of stroke risk after the index TIA within 2, 7, 30, and 90 days were, respectively, 2.5% (95% CI, 0.7-6.2), 5.6% (95% CI, 2.6-10.3), 6.2% (95% CI, 3.0-11.1), and 11.2% (95% CI, 6.8-17.1). ABCD2 score was strongly associated with stroke occurrence after index TIA: the areas under the receiver operating characteristic curve at 2, 7, 30, and 90 days were, respectively, 0.85 (95% CI, 0.72-0.97), 0.69 (95% CI, 0.56-0.82), 0.69 (95% CI, 0.56-0.85), and 0.76 (95% CI, 0.67-0.86). No patients with an ABCD2 score <4 had a stroke within the 90-day follow-up period. CONCLUSIONS: This study adds new data on TIA incidence and prognosis and it further validates the ability of the ABCD2 score to identify patients at early risk for stroke.