Not All Patients Should Be Admitted to the Hospital for Observation After a Transient Ischemic Attack
Should this patient be urgently evaluated? Yes, this patient should be evaluated because any transient ischemic attack (TIA), regardless of ABCD2 score,1 may be due to significant arterial stenosis or a cardiac source of embolism that needs to be urgently detected and treated.
When? This person should be evaluated within hours after the first call to medical attention because a devastating stroke may quickly follow a TIA. We therefore should be available 24/7. We have shown that this is feasible but needs good organization.2
Where? The best setting is the one that is effective. A TIA clinic can be located in the outpatient clinic of the stroke center, in the emergency department, or in a day hospital. In our model, we admit all patients with TIA to a dedicated TIA clinic located at the day hospital, which is a part of our comprehensive stroke center. We then triage patients based on actual findings with the diagnostic tests, not with the ABCD2 score.
Which diagnostic tests? In my opinion, MRI should be the preferred neuroimaging, although a default CT scan is also fine. Immediate arterial evaluation is mandatory. We perform ultrasound examination of extracranial arteries as well as transcranial Doppler. Several of our patients also have MR angiography or CT angiography. All patients undergo immediate cardiac evaluation including at least physical examination and an electrocardiogram. In case of abnormal findings, they undergo immediate transthoracic echocardiography and transesophageal echocardiography examinations. Patients with normal findings may have their cardiac evaluation within the next days (transthoracic echocardiography and transesophageal echocardiography, Holter electrocardiogram±transtelephonic electrocardiogram).
How long should the patient stay in the hospital? In our model, the work-up as described here is performed in <4 hours. Our published results show that only 25% of patients with TIA after this kind of triage need to be hospitalized >1 day (eg, for management of atrial fibrillation, intracranial hemodynamic compromise, and symptomatic carotid or intracranial stenosis); and 75% of the patients can be discharged home immediately with a prescription of secondary prevention treatment according to guidelines recommendations. This strategy proved to have a very low risk of stroke within 90 days. The overall risk was 1.24% for both hospitalized and immediately discharged patients; 3.18% when the analysis was restricted to patients presenting with an ABCD2 score >4; and 2.08% for those who were seen within 24 hours of symptom onset.2 This strategy also, of course, improves patients' satisfaction. This kind of triage based on actual findings in the patient with a “P” rather than on a score with a “S” is preferable when possible. I believe we should perform individualized medicine and that this is possible in all stroke centers provided a minimum of organization. In the past 2 decades, we have developed stroke units; now we should include dedicated TIA clinics in our stroke centers.
The National Institute for Health and Clinical Excellence (NICE) recommends that patients with TIA with a ABCD2 score ≥4 be evaluated and treated immediately at a TIA clinic or other facilities and that those with ABCD2 score <4 could be evaluated within 1 week.3 We have shown that among patients with ABCD2 score <4, 20% of patients have actual findings that need immediate medical decision-making and that these patients have a similar 90-day risk as patients with ABCD2 score ≥4.1,4 I believe that the NICE guidelines should be revised to require immediate electrocardiography and arterial imaging for all patients with TIA regardless of ABCD2 score.
Many consider that all patients with ABCD2 score ≥4 should be hospitalized for at least 1 day for “monitoring” in parallel to full diagnostic tests and starting secondary prevention treatment and that patients with ABCD2 score <4 should not be admitted. This strategy might lead to a loss of chance for these patients.1
Finally, some use the ABCD2 score as a triage tool after a quick triage based on actual findings with cardiac and arterial diagnostic tests to determine if the patient will be hospitalized for monitoring ≥24 hours.5 This is probably a safer use of ABCD2 score as a triage tool that needs to be evaluated.
So, another controversy is about patients with negative or not clinically relevant findings after quick triage based on diagnostic tests. Should we systematically admit all these patients with TIA to the hospital >1 day or should we immediately discharge them regardless of ABCD2 score? We found that the 90-day risk in patients with TIA with negative findings who were immediately discharged home after evaluation is reassuringly low: 1.28% overall (regardless of ABCD2 score), 3% if ABCD2 score was ≥4, and 0.4% if ABCD2 was score <4.4 Given this difference in the risks between those with negative findings and ABCD2 score ≥4 or <4, there is uncertainty as to whether those with ABCD2 score ≥4 should all be monitored in the hospital for >1 day. This certainly needs a randomized evaluation.
Sources of Funding
Supported by the SOS-ATTAQUE CEREBARALE Association (a not-for-profit stroke survivor and research organization).
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association. This article is Part 2 of a 3-part article. Parts 1 and 3 appear on pages 1446 and 1450, respectively.
- Received November 16, 2011.
- Revision received December 28, 2011.
- Accepted January 18, 2012.
- © 2012 American Heart Association, Inc.
- Amarenco P,
- Labreuche J,
- Lavallée PC,
- Meseguer E,
- Cabrejo L,
- Slaoui T,
- et al
National Institutes of Health and Clinical Excellence. NICE clinical guideline 68, Juy 23, 2008. Stroke diagnosis and initial management of acute stroke and transient ischaemic attack (TIA). Available at: www.nice.org.uk/nicemedia/live/12018/41331/41331.pdf.
- Amaenco P,
- Labreuche J,
- Lavallée PC
- Olivot JM,
- Wolford C,
- Castle J,
- Mlynash M,
- Schwartz NE,
- Lansberg MG,
- et al