TIA Patients:Risk Stratified & Prompt Action Prevent Strokes?

April 2008

TIA Patients – Can They Be Rapidly Risk Stratified & Does Prompt Action Prevent Strokes?

Search Strategy: Remembering that the Washington University Journal Club had discussed TIA at a past Journal Club, you first turn to the website archiving previous discussions (here). You then recall a five-item TIA clinical decision rule (CDR) derived by Dr. Johnston so you decide to perform a Web of Science search to determine whether the rule has been modified, validated, or replaced. Your search reveals that as of January 17, 2008 Johnston’s article had been cited 251 times including the development of a rival CDR called the ABCD score (http://pmid.us/15993230) which has since been merged with Johnston’s rule into yet a third CDR – the ABCD2 score (below)! You had planned to perform a PUBMED clinical query for TIA therapy, but fortuitously your initial Web of Science search also uncovered several promising rapid-therapy for TIA interventions which you decide to explore further.

Mr. S., a 67 year old left-handed male, presents to your ED complaining of a now-resolved one-hour episode of left facial droop and left arm weakness. He has never had anything like this happen before. He is a smoker with a history of hypertension and diabetes.

On exam the patient is awake, alert, and appropriate. His blood pressure is 130/90, heart rate 80 (sinus), respiratory rate 14, and he is afebrile with 96% oxygen saturation on room air. The rest of his physical exam was normal. His EKG is unremarkable, but he has no old EKG for comparison. A head CT reveals nothing abnormal.

Concerned about a transient ischemic attack (“angina of the brain”), you consult your Neurologist to admit Mr. S. Your Neurologist feels that with symptom resolution, patients like Mr. S. are safe for outpatient management with “about a 15% one-year stroke risk based on the literature”. Your Neurologist is unable to provide you with any TIA-risk stratification tools and you are uncomfortable with Mr. S. waiting at home for his cardiac Echo and carotid Dopplers. Your Neurologist colleague challenges you with the statement, “Has any study demonstrated an improved outcome among TIA patients admitted from the ED rather than evaluated as an outpatient?” Your ire raised, you turn to the medical literature.

PICO Question #1

Population: ED TIA patients

Intervention: Risk stratification tools

Comparison: Unaided clinical intuition (gestalt)

Outcome: Stroke, death

PICO Question #2

Population: ED TIA patients

Intervention: Admission for diagnostic testing

Comparison: Outpatient diagnostic testing and follow-up

Outcome: Stroke, death


First years: Risk of Stroke Early After Transient Ischaemic Attack: A Systematic Review and Meta-Analysis, Lancet Neurology 2007; 6: 1063-1072 (http://pmid.us/17993293)

Second years: Validation and Refinement of Scores to Predict Very Early Stroke Risk After Transient Ischaemic Attack, Lancet 2007; 369: 283-292. (http://pmid.us/17258668)

Third years: An Emergency Department Diagnostic Protocol for Patients with Transient Ischemic Attack: A Randomized Controlled Trial, Ann Emerg Med 2007; 50: 109-119. (http://pmid.us/17490788)

Fourth years: A Transient Ischaemic Attack Clinic with Round-the-Clock Access (SOS-TIA): Feasibility and Effects. Lancet Neurology 2007; 6: 953-960. (http://pmid.us/17928270)


Article 1: Prospective Risk of stroke early after transient ischaemic attack: a systematic review and meta-analysis, Lancet 2007; 6: 1063-1072

Article 2: Validation and refinement of scores to predict very early stroke risk after Transient Ischaemic Attach, Lancet 2007; 369:283-292

Article 3: An Emergency Department Diagnostic Protocol for Patients with Transient Ischemic Attack: A Randomized Controlled Trial, Annals of EM 2007; 50:109-119

Article 4: A transient ischaemic attack clinic with round-the-clock access (SOS-TIA): feasibility and effects, Lancet-Neurology 2007; 6: 953-960

Bottom Line

In assessing 2-day post-TIA stroke risk, the California rule, ABCD, and ABCD2 display similar prognostic capabilities. The ABCD2 prognostic guide may be slightly superior. To minimize post-TIA stroke risk among ED patients a score of ≤ 1 was associated with a 0% 2-day risk on the validation cohorts. ABCD2 score of ≤ 3 has a 1% 2-day stroke risk while a score of > 5 has an 8% risk. Before widespread adoption of the ABCD2 rule future research should validate the prognostic test characteristics prospectively on distinct populations. Furthermore, clinicians will need to be convinced that admission of high-risk TIA patients improves outcomes.

Based upon a single center ED-observation based model, TIA evaluation may be enhanced with more frequent best-evidence diagnostic testing at substantially reduced expense and length-of-stay. Another French-based TIA clinic available around-the-clock reduced post-TIA and post-minor stroke 90-day and 1-yr stroke rates compared with historical cohorts with NNT=21, although the interventions which resulted in these reductions are not clear. Future ED-based TIA protocol research should expand outcomes to include interventional post-TIA stroke reduction and application of the rapid protocol to high-risk patients using the ABCD2 prognostic tool.