J Pak Med Assoc. 2026 Aug;76(8):1277-1282. doi: 10.47391/JPMA.30130.
ABSTRACT
OBJECTIVE: To investigate the accuracy of patient demographic data and non-contrast computed tomography brain scan findings in predicting the vascular aetiology of non-traumatic intracranial haemorrhage.
METHODS: The retrospective cohort study was conducted from April 15 to October 15, 2023, at Armed Forces institute of radiology and imaging (AFIRI) Rawalpindi, Pakistan, and comprised data from May 1, 2022, to September 30, 2023, of patients aged >18 years who presented with non-traumatic intracranial haemorrhage and underwent computed tomography angiography for the detection of underlying vascular lesions. Non-contrast computed tomography criterion was used to predict the presence of vascular aetiologies in patients of non-traumatic spontaneous intracranial haemorrhage. Independent predictors of vascular aetiology were identified, and the vascular intracerebral haemorrhage scoring system was crafted which was internally validated. Data was analysed using SPSS 25.
RESULTS: Of the 237 patients with mean age 55.9±14.5 years (range: 18-85 years), 136(56%) were male and 101(42%) were female. Vascular aetiology was confirmed in 153(64.5%) patients. Independent predictors included age <55 years (odds ratio: 2.94, 95% confidence interval: 1.48-5.83), lobar location (odds ratio: 2.89, 95% confidence interval: 1.47-5.67), absence of hypertension (odds ratio: 5.12, 95% confidence interval: 2.67-9.82), and perilesional oedema (odds ratio: 2.76, 95% confidence interval: 1.34-5.67). The vascular intracerebral haemorrhage score ranged 0-8, and showed good discrimination (area under the receiver operating characteristic curve: 0.81, 95% confidence interval: 0.75-0.87). A cut-off value of ≥4 yielded sensitivity 82%, specificity 74%, and Youden's index 0.56.
CONCLUSIONS: The vascular intracerebral haemorrhage score demonstrated effectiveness in anticipating the presence of vascular causes. After external validation, the scoring system may assist in triaging patients for computed tomography angiography.
PMID:42703117 | DOI:10.47391/JPMA.30130