Abstract Number: Esoc2026ot104 Prevalence and Clinical Characteristics of Stroke Mimics Among Patients With Suspected Acute Stroke in the Emergency

Abstract Background and aims Stroke mimics pose diagnostic challenges in emergency departments (ED). We aimed to determine the prevalence, aetiologies, and clinical predictors of mimics among suspected acute stroke patients. Methods Objective To determine the prevalence, aetiologies, and clinical predictors of stroke mimics among patients with suspected acute stroke within 9 hours of symptom onset in the ED. Results Methods A retrospective cohort study was conducted involving 3,170 adult patients presenting to the emergency department of Srinagarind Hospital between January 2020 and December 2024 and were triaged through the institutional Stroke Fast-Track protocol. Data extracted from electronic medical records included demographics, comorbidities, presenting symptoms, and neuroimaging findings. Final diagnoses of stroke mimics were determined by attending neurologists. Statistical analyses were performed using R software, including descriptive statistics and logistic regression to identify symptom predictors of stroke mimics. Conclusions Results Of 3,170 patients, 564 (17.8%) were diagnosed with stroke mimics. Seizure-related conditions (16.3%) were the most common aetiology, followed by cardiovascular non-stroke causes (11.7%) and infectious or inflammatory disorders (11.2%). Other mimics included metabolic disturbances, vestibular syndromes, migraine, structural brain lesions, and psychogenic disorders. Univariate analysis identified symptoms significantly associated with stroke mimics, including seizure (OR 4.96; 95% CI 3.40–7.20), memory disturbance (OR 3.59; 95% CI 1.70–7.41),and fever (OR 3.35; 95% CI 2.11–5.51). Conversely, classic focal deficits—motor weakness (OR 0.45), dysarthria (OR 0.59), and facial palsy (OR 0.79)—were negatively associated with mimics, indicating a higher likelihood of true stroke. Conflict of interest