BRAIN COMPUTED TOMOGRAPHIC FINDINGS OF STROKE PATIENTS IN MAIDUGURI, NORTH-EASTERN NIGERIA

Background: Stroke is a major neurological condition with high disability and mortality rates in developing countries including Nigeria. Computed tomography (CT) scan is an important imaging modality in confirming the diagnosis of stroke, classifying stroke subtypes and ruling out stroke mimics. Objectives: To determine the CT findings of patients with stroke in Maiduguri, North-eastern Nigeria. Methodology: A retrospective hospital-based study conducted at the Department of Radiology, Federal Neuro-psychiatric Hospital, Maiduguri, North-eastern Nigeria over four-year period from January 2015 to December 2018. All patients were clinically diagnosed with stroke and referred to the department for a brain CT scan. The request forms and CT reports were retrieved from the Archive and analysed. All patients were examined with a 16-slice General Electric CT scanner. Data were analysed using SPSS statistical software version 23. Results: There were 138 patients and out of which, 83 (60%) were males and 55 (40%) females with a male to female ratio of 1.5:1. The mean age was 51.6 ± 16.6 years with an age range of 5-85 years and the majority of the patients were middle-aged group. The CT findings showed 88 (64%) were confirmed stroke cases and the remaining 50 (36%) were misdiagnoses of stroke, with normal finding in 23% and stroke mimics in the remaining 13%. Brain infarction was the most common stroke subtype accounting for 64% of the patients, followed by intracerebral haemorrhage of 33% and 3% had subarachnoid haemorrhage. The left cerebral hemisphere was the most commonly involved. Cerebral atrophy and brain tumours were the two most common stroke mimics in this study and they were mostly seen in females than males. Conclusion: In this study, the accuracy rate of clinical diagnosis of stroke was low. Therefore, CT scan is the 'gold standard’ in the diagnosis of stroke and it is strongly recommended in the management of all stroke patients.