Abstract

Background: Since the 7-day stroke risk in DWI-normal TIA is negligibly low (0% to 0.4%), approximately 20 times lower than the corresponding risk after DWI-positive TIA, a clinically relevant question in the emergency setting is which patient with the clinical syndrome consistent with a TIA is likely to have MRI evidence of acute ischemia and should receive urgent imaging. This question has cost and management implications. We sought to determine the relationship between clinical TIA characteristics and the probability of detecting acute infarction on DWI in patients admitted to the emergency department with focal neurological symptoms lasting <24 hours. Methods: This was a retrospective, single center, emergency-based study of 586 consecutive patients. We explored the relationship between DWI positivity and the following clinical TIA characteristics: age, gender, symptom duration, symptom type, vascular risk factors, ictal headache, prior history of stroke or TIA, ABCD 2 score, and the type of spell (objective vs. subjective). Results: Univariate predictors of infarction on DWI included speech disturbance (OR=1.7, 95%CI 1.2-2.5), weakness (OR=2.3, 95%CI 1.6-3.3), visual symptoms (OR=0.4, 95%CI 0.2-0.7), isolated sensory symptoms (OR=0.3, 95%CI 0.2-0.60), subjective symptom (presenting with focal symptoms other than weakness or speech disturbance, OR=0.4, 95%CI 0.3-0.6), and ABCD 2 score≥4 (OR=1.6 95%CI 1.1-2.20). The positive predictive value for diagnosis of acute infarction on DWI was 38% for speech disturbance, 43% for weakness, 17% for visual symptoms, 16% for isolated sensory symptoms, 20% for subjective symptom, and 27% for ABCD 2 score≥4. Recursive partitioning analysis revealed a regression tree based on weakness, isolated sensory symptoms, and visual symptoms. According to this tree, the lowest probability for a given TIA patient to have a positive DWI was 17%. Conclusions: DWI provides diagnostic information in all subsets of TIA irrespective of clinical TIA characteristics. DWI is a valuable tool in the emergency management of TIA and offers potential to lead to improved TIA outcome particularly in settings where triage is based on clinical features.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call