Abstract
Background Emergency department (ED) physicians often encounter patients presenting symptoms related to neurological disorders. However, due to the variation in imaging modalities, it is difficult for physicians to ascertain the most appropriate modality for diagnosis. Therefore, the present meta-analysis was designed to evaluate the effectiveness of various neuroimaging modalities in the detection of acute neurological disorders, with a focus on acute stroke and subarachnoid hemorrhage (SAH).Methods PubMed, Web of Science, Cochrane Library, and Google Scholar databases were searched for records published until May 2024. Studies published in English and evaluating patients presented for emergency assessment of symptoms related to stroke or SAH were included. Moreover, Quality assessment was performed using the QUADAS-2, and statistical analyses were performed with STATA version 16 software.Results 19 studies (14 focused on acute stroke and 5 on SAH) were included in the final analysis. The pooled analysis showed that Diffusion-weighted image (DWI), computed tomography perfusion (CTP), non-contrast computed tomography (NCCT), and conventional magnetic resonance imaging (MRI) had an overall sensitivity of 91%, 68%, 44%, and 61%, and an overall specificity of 93%, 91%, 90%, and 97%, respectively in the diagnosis of acute stroke. On the other hand, the pooled analysis indicated that CT has an overall sensitivity and specificity of 95% and 100% in diagnosing SAH.Conclusion DWI was more sensitive than NCCT, conventional MRI, and CTP in diagnosing acute stroke. Furthermore, CT was highly sensitive in diagnosing SAH, especially when performed within 6 hours of symptom onset. However, lumbar puncture is still required after a negative CT because there have been a few false negatives.