Sensitivity, Specificity, Predictive Values and Accuracy of clinical diagnosis of Acute Stroke

Abstract

Background: Stroke is a leading cause of death and disability globally and particularly in low and middle-income countries, and the burden is increasing. To prevent complications and permanent defects in stroke, early diagnosis is the key that can easily obtained by a CT scan of brain. However, quick access to CT scanning is not available in every country and hospital specially in Bangladesh, various clinical findings especially neurological signs and symptoms and risk factors differentiation are helpful in differentiating the types of stroke. Objective: This study aimed to see the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of clinical diagnosis in the detection of stroke subtype. Methods: This hospital based cross-sectional comparative study was conducted in Department of Medicine, Rangpur medical college hospital, Rangpur, Bangladesh from January 2010 to December 2011 on three hundred (300) suspected acute stroke patients selected by purposive sampling technique. The clinical diagnosis of type of stroke was made on the basis of mode of presentation, risk factors and signs and confirmed by CT scan of brain within 1 week of attack. Then the clinical diagnosis was compared with the results of CT scan. Statistical analyses related with this study were performed by using of SPSS-19 package program. Results: Among the 300 patients, 73(24.3%) patients were clinically diagnosed as hemorrhagic stroke and 227(75.6%) patients were as infarctive stroke. Out of 73 clinically diagnosed haemorrhagic stroke patients, CT scan revealed that 61 (83.6%) patients had intracerebral hemorrhage, 5 (6.8%) had infarct. And out of 133 diagnosed ischemic stroke patients, CT scan revealed that 203 (89.4%) patients had infarction, 6 (2.6%) had intracerebral hemorrhage. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of clinical diagnosis of hemorrhagic stroke were 91.0%, 94.8%, 83.6%, 97.4% and 94.0% respectively. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of clinical diagnosis of infarctive stroke were 97.6%, 73.9%, 89.4%, 92.2% and 90.3% respectively. Conclusion: CT scan of brain remains the gold standard for differential diagnosis, but the availability of CT scan is not always feasible, and it is virtually impossible to submit all stroke patients to CT scan. Adequate knowledge on risk factors, clinical features and initial investigations may contribute to such a differentiation of cerebral infarction from intracerebral hemorrhage with high accuracy where rapid access to Computed Tomography (CT) is lacking. J Rang Med Col. March 2023; Vol. 8, No. 2:33-39