Paper
TIAs
Published May 1, 1988 · L. Caplan
Neurology
63
Citations
0
Influential Citations
Abstract
“Transient ischemic attack” (TIA) is a familiar term now very well known to practitioners, house officers, and students alike. Even patients frequently call their doctors, worried that they have had a “TIA.” In this age of increasing media-centricity, we can expect even more hype and more patient and physician awareness of TIA and other putative “stroke symptoms.” Yet the term and concept of TIA are relatively new. We owe to Miller Fisher the first detailed description and review of the phenomenology of TIAs, described as “prodromal k e t ing attacks of paralysis, numbness, tingling, speechlessness, unilateral blindness or dizziness,” which nearly always preceded cerebral infarction in his patients with occlusion of the internal carotid artery.’ This work was stimulated by Dr. Fisher’s experience with a single patient who had episodes of transient right monocular blindness preceding a left hemiplegia. The patient remarked, “Isn’t it funny that I went blind in the wrong eye?”2 Previously, Peabody3 and Oslefl had each described, near the turn of the twentieth century, patients with multiple TIAs. Most textbooks of medicine and neurology written before 1950 did not mention TIAs, but Gowers referred to “premonitory symptoms” before brain softening: and Gordone described a “preapoplectic period” during which patients had headache and other ill-defined symptoms before their strokes. In this issue of Neurology, two articles7s8 about TIAs both conclude: (1) TIAs are usually very short; the majority of episodes clear within 1 hour, and most patients with signs lasting 4 hours or longer develop strokes; and (2) some patients with clinical TIAs have infarcts shown on CT in locations appropriate to their transient symptoms. The finding of infarction by CT or other neuroimaging techniques is important and conclusively links TIA with cerebral infarction. Although Levy7 does not note the CT results of his patients, he does review reports that showed a range between 25% and 50% of positive CTs among four series of TIA patients. Werdelen and Juhlel.8 noted CT-documented infarcts in three of ten TIA patients. In addition to reports cited by these a number of others have also studied the incidence and location of ischemic lesions shown by CT and other imaging tests, such as magnetic resonance imaging and positron emission tomography, in patients with attacks meeting the definition criteria for TIAs. Nicolaides et ale studied 149 patients with hemispheric TIAs, 48% had an infarct on CT and 35% of the patients had an infarct in the symptomatic hemisphere. Fortythree percent of 70 patients with transient monocular blindness but no history of stroke or cerebral symptoms had infarcts on CT? Nicolaides et al analyzed the results of 17 studies of the frequency of infarcts on CT in patients with hemispheric TIAs. Of 783 patients, 154 (20%) had infarction on CT. Of 371 patients analyzed for the side of symptoms, 11% had infarcts in the symptomatic hemi~phere.~ MRI is clearly more sensitive than CT in documenting cerebral infarcts. In one study,’O seven of 22 patients with TIAs (32%) showed infarcts on CT (six in the symptomatic hemisphere) while 17 (77%) had infarcts on MRI (15 in the symptomatic hemisphere). Salgado et all1 also studied TIA patients with both CT and MRI. Of 21 patients with TIA or amaurosis fugax (two had a clinical stroke in the same vascular temtory as their TIA), eight CTs (42%) and 16 MRIs (84%) showed focal ischemia.” Only 50% of the CT findings correlated with symptoms, but 81% of the abnormal MRIs had lesions appropriate to the symptoms. KinkleI2 found that 13 of 60 TIA patients (22%) had focal CT abnormalities, while 42 (70%) had MRI lesions in periventricular (12), cortical-subcortical (18), and watershed (12) distribution. Emission tomography13 and positron emission tomography also show a high incidence of abnormality in TIA patients.14-ls The cerebral metabolic rate for oxygen and the oxygen extraction fraction are often altered in the symptomatic region long after TIAs have resolved. The imaging data are very convincing that many patients with TIA have persistent, possibly reversible, brain tissue dysfunction, and probably at least one in four have a cerebral infarct. We do not know why symptoms are only transient in the patients with infarction. The finding that most TIAs are short is not new. In fact, we do not know of any study that has found otherwise. In the largest single study of TIAs, the NINCDSfunded Cooperative Study of Transient Ischemic Att ack~ , ’~ 1,307 patients were analyzed; the median duration of attacks was 14 minutes in patients with “carotid artery symptoms” and 8 minutes in the “vertebrobasilar system” symptom group. Ninety percent of hemispheral attacks lasted less than 6 hours, and 90% of
TIAs are short episodes of symptoms that often clear within 1 hour, and infarcts on CT or other neuroimaging techniques link TIAs to cerebral infarction.
Full text analysis coming soon...