RESUMO
Background: The gold standard for distinguishing stroke subtype is non-contrast CT. However, it's still prohibitively expensive and out of reach for the majority of recourse-constrained settings. Clinically, not all patients will have a definite diagnosis of hemorrhagic/ischemic stroke. To overcome these challenges and improve clinical bedside diagnosis, clinical stroke scores for stroke subtypes have been developed and recommended to be used in the absence of appropriate imaging modality. Methods: We conducted a prospective cross-sectional study among stroke patients to compare the accuracy of level of clinical stroke score methods in differentiating stroke type with CT. it was conducted on 140 people at MTU teaching hospital in Bench-Sheko Zone, South-west Ethiopia. Data were collected using check list. Analysis of the data was done using SPSS version 24. Results: Our result revealed an incidence of hemorrhagic stroke were 50%, ischemic stroke were 48.6% by CT evaluation. Specificity, sensitivity, positive predictive value, negative predictive value and the overall accuracy of Siriraj stroke score for differentiation of hemorrhage from ischemic stroke were 68.6%, 83.9%, 74.6%, 79.5%, and 82% respectively, the Guys score were 89.7%, 47.8%, 73.3%, 74.5% and 74.5% respectively and while the Bensson score were 88.6%, 35.3%, 75%, 58.5%, and 62.3% respectively. Conclusion: We conclude that Siriraj stroke score showed good sensitivity and fair overall accuracy for hemorrhagic stroke even if it had poor specificity.