Predictive implications of recurrent transient ischemic attacks in large-artery atherosclerosis.

BACKGROUND: It is uncertain whether recurrent transient ischemic attacks (R-TIAs), when comparing with single TIAs (S-TIAs), have any distinct mechanisms. METHODS: All consecutive patients with TIAs, who had been admitted for a 2-year period, were divided into two groups: those who had R-TIAs and those who had S-TIAs. Registry data, medical records, and imaging findings were reviewed and compared between the two groups. RESULTS: There were 85 patients who had TIAs: 42 patients had R-TIAs, and 43 patients had S-TIAs. On univariate analysis, R-TIA patients had less cardiac embolic TIA sources, less weakness, less speech disturbances, shorter symptom duration, a longer time interval from onset to treatment, less abnormalities on diffusion-weighted magnetic resonance imaging, and more significant relevant arterial stenoses. After logistic regression analysis, independent factors associated with R-TIAs were symptom duration < 10 min (odds ratio OR 3.62; 95% confidence interval CI 1.37-9.57), > or = 50% stenosis of the clinically relevant artery (OR 7.08; 95% CI 1.29-38.71), and absence of cardiac embolic sources (OR 0.04; 95% CI 0.002-0.71). CONCLUSIONS: R-TIAs may have pathophysiological mechanisms distinct from those of S-TIAs and so may provide a clue for the etiologic diagnosis, in that patients with R-TIAs are more likely to have large-artery atherosclerosis.