# Distinguishing stroke from transient ischemic attack using plaque characteristics and arterial transit artifacts

## TL;DR

This study shows that combining high-resolution MRI and arterial spin labeling can effectively distinguish between stroke and transient ischemic attack by analyzing plaque features and arterial transit artifacts.

## Contribution

The study introduces a combined imaging model using plaque characteristics and arterial transit artifacts to improve stroke-TIA differentiation.

## Key findings

- Significant differences in plaque features and arterial transit artifacts were found between stroke and TIA groups.
- The combined HRMRI and ASL model achieved high diagnostic accuracy (AUC 0.926) for distinguishing stroke and TIA.
- ASL offers a simpler alternative to HRMRI for evaluating ischemic stroke with comparable diagnostic performance.

## Abstract

We aimed to investigate the differences in plaque characteristics and hemodynamics in patients with ischemic stroke and transient ischemic attack (TIA), comparing the diagnostic abilities of high-resolution magnetic resonance imaging (HRMRI) and arterial spin labeling (ASL) for ischemic stroke.

This retrospective analysis included patients who underwent HRMRI and ASL between October 2020 and December 2023. We compared clinical risk factors, vascular plaque characteristics, and the presence of arterial transit artifacts (ATAs) at post-labeling delays (PLDs) of 1.5-s and 2.5-s between stroke and TIA groups. Multivariate logistic regression analysis was used to evaluate the diagnostic performance of different prediction models combining clinical factors, differential plaque characteristics, and the presence of ^PLD ATAs.

A total of 147 patients (mean age, 57.12 ± 13.08 years; 102 men) were initially included in this study, divided into stroke (79) and TIA (68) groups. Significant differences in vascular positive remodeling, intraplaque hemorrhage, enhancement ratio, and the presence of 1.5-s and 2.5-s ATAs (p < 0.05) were observed between groups. Combined HRMRI and ASL performed best in distinguishing ischemic stroke and TIA (area under the curve [AUC], 0.926; 95% confidence interval [CI], 0.885–0.967), with no significant difference in ischemic stroke diagnostic performance between HRMRI and ASL (95% CI, −0.039 to 0.087, Z = 0.742, p = 0.458).

A model combined with plaque characteristics and ATAs showed good diagnostic performance in distinguishing between TIA and stroke in patients with intracranial atherosclerotic stenosis. ASL provides a simpler imaging evaluation method than HRMRI, and ATA evaluation may become a more widely used imaging marker in clinical practice.