What question did this study set out to answer?

The study aims to identify factors that can differentiate stroke mimics from acute ischemic stroke and transient ischemic attack.

February 2, 2026

Abstract DP252: Predicting Stroke Mimics: A Clinical–Radiological Model for Early Differentiation from Acute Ischemic Stroke and Transient Ischemic Attack

Key Points

The study aims to identify factors that can differentiate stroke mimics from acute ischemic stroke and transient ischemic attack.
Conducted a prospective multicenter observational study
Collected data on demographics, medical history, and clinical parameters
Used multiple logistic regression to develop predictive models
Established stroke mimics based on clinical course and follow-up neuroimaging
Analyzed both demographic/clinical variables and neuroimaging results
1071 patients enrolled; 699 had complete datasets
Identified 457 acute ischemic strokes, 65 transient ischemic attacks, and 177 stroke mimics
Model 1 found elevated INR, history of epilepsy, and absence of atrial fibrillation as predictors of stroke mimics
Model 2 added younger age and specific negative neuroimaging results as additional predictors

Abstract

Background: Recognition of acute stroke in the emergency setting remains a major diagnostic challenge. Up to one quarter of patients admitted with suspected stroke are ultimately diagnosed with a stroke mimic (SM). Failure to distinguish SM from cerebrovascular events may result in inappropriate thrombolysis, delayed initiation of targeted therapy, and inefficient use of healthcare resources. Objective: To identify independent predictive factors, including clinical history, examination findings, and radiological multimodal CT parameters, that support differentiation of SM from AIS and TIA. Methods: We conducted a prospective multicenter observational study with patients admitted with suspected acute stroke to centers between June 2023 and May 2024. SM were established by stroke experts based on clinical course and follow-up neuroimaging. Collected variables included demographics, medical history (epilepsy, migraine, psychiatric illness, cognitive disorder, hypertension, diabetes mellitus - DM, atrial fibrilation - AF) and admission clinical parameters (BP, INR, blood glucose, NIHSS). Neuroimaging assessment comprised NCCT, CTA, and CTP (Brainomix e-Stroke v10.1). Exploratory analysis used multiple logistic regression. Two models were developed: Model 1 included demographic, anamnestic, and clinical variables as covariates; Model 2 additionally incorporated early neuroimaging. Results: A total of 1071 patients were enrolled; 699 had complete datasets (median age 75 years; 45% women). Final diagnoses: 457 AIS (65.4%) , 65 TIA (9.3%) , and 177 SM (25.3%), most often seizures (31.6% of all SM) . In Model 1 identified higher INR (OR 4.55, 95%CI 2.04–10.40, p<0.001), history of epilepsy (OR 5.86, 95%CI 2.76–12.95, p<0.001), and absence of AF (OR 0.51, 95%CI 0.30–0.83, p=0.008) as significant predictors of SM. In Model 2 additional predictors included younger age (OR 0.97, 95% CI: 0.95–0.99, p=0.006), higher NIHSS (OR 1.09, 95% CI: 1.05–1.14, p<0.001), neg CTA (OR 8.81, 95% CI: 4.08–21.19, p<0.001), and neg CTP (OR 18.42, 95% CI: 7.90–48.79, p<0.001) as significant factors associated with SM. Conclusion: Our exploratory analysis identified clinical factors (history of epilepsy, elevated INR, absence of AF) and radiological findings (neg CTA and CTP) as independent predictors of SM. As this project is ongoing, the next step is to validate a robust predictive model in independent cohorts before implementation.

Read Full Paperexternally

Bookmark

View Full Paper

Cite This Study

Bar et al. (Thu,) studied this question.

synapsesocial.com/papers/6980fbbec1c9540dea80d877 https://doi.org/https://doi.org/10.1161/str.57.suppl_1.dp252

Bookmark

View Full Paper