To determine the outcome of patients with carotid transient ischemic attacks (TIAs) and normal cerebral angiograms, we assessed 68 patients (40 men, 28 women) aged 24-72 (mean 53.5) years for recurrent TIAs and strokes and for the development of cardiac disease over 2-6 (mean 4.4) years. All but one patient had a follow-up interview in early 1987; that patient had died of an unrelated cause (lung cancer) 18 months after the presenting TIA. The diagnosis was changed at the follow-up interview in three patients (multiple sclerosis, meningioma, migraine). Among the 64 remaining patients, at admission cranial computed tomography had shown cerebral infarction in 11 of 64, two-dimensional echocardiography had been abnormal in nine of 61, Holter monitoring had been abnormal in eight of 45, and twelve-lead electrocardiography had been abnormal in three of 64. Two patients had abnormalities on both echocardiography and Holter monitoring. At the follow-up interview of the 64 remaining patients, TIAs had recurred in nine and three had developed a completed stroke; cardiac disease (angina in seven, myocardial infarction in four) was noted in 11 patients. Findings from cardiac investigations on admission in the nine patients with recurrent TIAs had been abnormal in six and normal in three; all three patients who developed a stroke had had abnormal cardiac findings. Overall, further neurologic or cardiac events occurred in 12 of 46 patients (26%) with normal and in 10 of 18 patients (55.5%) with abnormal findings on admission (/?<0.01). In the presence of normal angiograms, extensive cardiac investigations may help predict the outcome of patients with TIAs. (Stroke 1988;19:1223-1228) C erebral angiography shows atherosclerosis in the majority of patients who have transient ischemic attacks (TIAs) in the carotid artery territory,'-3 with the subsequent risk of stroke over 5 years ranging from 10% to 43%. 4 -9 Angiograms may, however, be normal in approximately 25% of these patients. 10 The reported frequency of normal angiograms in patients with TIAs has varied from as low as 7%" to as high as 40%. 12 The etiology, natural history, and prognosis in these patients is not well defined. Some articles suggest an excellent prognosis 13 while others report an outcome similar to that of patients in whom angiography shows atherosclerotic narrowing."14 However, few patients were studied, 11 angiography was incomplete, 1415 and follow-up was limited
13; there- Received April 4, 1988; accepted May 24, 1988. fore, no definite conclusions can be drawn from these studies. Our study was undertaken to evaluate the outcome of patients admitted over a 4-year period to University Hospital, London, Ontario, Canada, with TIAs and normal angiograms in whom cardiac investigations were extensive. Our results suggest that patients with cardiac disease (previously known or newly diagnosed during evaluation) have a higher incidence of subsequent cardiac and noncardiac complications. Cardiac investigations separate patients with TIAs and normal angiog...