Current issues of ACP Journal Club are published in Annals of Internal Medicine


Cardiac events after atypical transient ischemic attack

ACP J Club. 1993 Jan-Feb;118:24. doi:10.7326/ACPJC-1993-118-1-024

Source Citation

Koudstaal PJ, Algra A, Pop GA, et al, for the Dutch TIA Study Group. Risk of cardiac events in atypical transient ischaemic attack or minor stroke. Lancet. Sep



To compare the frequency of subsequent cardiac events in patients with typical and atypical symptoms of presumed transient ischemic attack (TIA) and minor ischemic stroke.


Cohort study with mean follow-up of 2.6 years.


Multicenter Dutch TIA Study.


997 patients who had had ≥ 1 TIA (symptoms reversible within 24 h) and 2130 who had had ≥ 1 minor stroke (patients had symptoms for longer than 24 h but remained independent in most activities of daily living). Patients with a presumed source of cardiac embolic event, vasculitis, or blood dyscrasias were excluded. Follow-up was 100%.

Assessment of risk factors

Patients completed a multiple-choice symptom checklist including onset, duration, nature, and number of attacks. Atypical symptoms included ≥ 1 of disturbances of vision in 1 or both eyes; tired or heavy sensation in ≥ 1 limb; sensory symptoms alone or a gradual spread of sensory symptoms; isolated disorder of swallowing or articulation, double vision, dizziness, or uncoordinated movements; accompanying symptoms of unconsciousness, limb jerking, tingling, disorientation, or amnesia. Computed tomography (CT) and electrocardiograms were done.

Main outcome measures

Vascular death (sudden cardiac death, death from stroke, myocardial infarction, congestive heart failure, or systemic bleeding); nonfatal stroke with symptoms persisting for > 24 hours and evidence from CT or an increase in disability; myocardial infarction. Patients were evaluated every 4 months by a neurologist. All events were reviewed independently by ≥ 3 assessors.

Main results

The 572 patients who had had atypical symptoms did not differ in overall incidence of vascular (cardiac and other) events from the 2555 patients who had had typical symptoms alone (about 15% in each group; hazard ratio [HR] 1.0 95% CI 0.8 to 1.2). The patients with atypical symptoms had higher rates of cardiac death alone (6% vs. 4%; HR 1.5, CI 1.1 to 2.3) and cardiac death or nonfatal myocardial infarction (8% vs. 6%; HR 1.4, CI 1.0 to 2.0) but lower rates of fatal or nonfatal stroke (6% vs. 9%; HR 0.6, CI 0.4 to 0.9).


Patients with atypical symptoms of transient ischemic attack or minor stroke had a lower risk for subsequent stroke but higher risk for cardiac death than patients with typical symptoms. This may be because the atypical symptoms resulted from primary cardiac events.

Source of funding: Not stated.

For article reprint: Dr. P.J. Koudstaal,Department of Neurology, University Hospital Rotterdam Dijkzigt, 40 Dr Molewaterplein, 3015 GD Rotterdam, Netherlands. FAX 31-10-463-5305.


A TIA or minor stroke is usually defined by its clinical features: the sudden onset of symptoms of focal neurologic dysfunction that are reversible and may be of vascular origin. The importance of this symptom complex is that it has etiologic, prognostic, and therapeutic relevance: It is usually caused by cerebral or retinal arterial occlusion resulting from atherothromboembolism, lipohyalinosis, or cardiogenic embolism; it is usually associated with an increased risk for stroke that can be reduced by controlling vascular risk factors, long-term antiplatelet therapy, and, if appropriate, carotid endarterectomy.

The interobserver variation in the clinical diagnosis of TIAs is, however, considerable. This is mainly because the diagnosis depends on interpreting symptoms—no diagnostic test for TIA exists. One neurologist's TIA is, therefore, not necessarily another neurologist's TIA. Koudstaal and colleagues have used the internationally accepted criteria to classify TIAs into "typical" and "atypical" TIAs. They found that typical TIAs are associated with an increased risk for stroke and atypical TIAs with an increased risk for cardiac events.

The implications of the study are that patients with TIAs and minor stroke are a heterogeneous group in terms of clinical features and prognosis, and, therefore, probable etiology and response to treatment. Patients with typical TIAs are likely to have cardiac or arterial thromboembolism as the underlying cause and an increased risk for stroke that may be reduced by treatments such as carotid endarterectomy. On the other hand, patients with atypical TIAs are more likely to have cardiac disease (i.e., cardiac arrhythmia) as the underlying cause and an increased risk for cardiac events, which is not likely to be reduced by carotid endarterectomy but which may be reduced by appropriate cardiac investigations and management in addition to control of vascular risk factors and long-term antiplatelet therapy.

The generalizability of these findings is dependent on a comparable precision of clinical diagnosis and the degree of coexisting cerebral atherothromboembolism and primary cardiac disease.

Graeme Hankey, MBBS
Royal Perth Hospital Perth, Western Australia