Poster Presentation Australian and New Zealand Stroke Organisation Conference 2025

Role of Computerized Tomography Perfusion in identification of Cerebral Venous Thrombosis in emergency department (#101)

Luis Mena Romo 1 , Octavio Garcia Silva 2 , Beng L. Alvin Chew 2 , Md Golam Hasnain 1 3 , James Thomas 2 4 , Afshin Borhani Haghighi 5 , Cecilia Ostman 1 , Mark W. Parsons 1 2 3 4 6 , Neil Spratt 1 2 3 , Carlos Garcia‐Esperon 1 2 3 6
  1. HUNTER MEDICAL RESEARCH INSTITUTE, UNIVERSITY OF NEWCASTLE, NEWCASTLE, NSW, Australia
  2. NEUROLOGY, JOHN HUNTER HOSPITAL, NEWCASTLE, NSW, Australia
  3. College of Health, Medicine and Wellbeing, University of Newcastle, Newcastle, NSW, Australia
  4. Departament of Neurology, Liverpool Hospital, Sydney, NSW, Australia
  5. Shiraz University of Medical Sciences, University of Shiraz, Shiraz, Iran
  6. FACULTY OF MEDICINE, UNIVERSITY OF NEWCASTLE, NEWCASTLE, NSW, Australia

Background/Aims
Cerebral Venous Thrombosis (CVT) patients often present with stroke-like symptoms. Increasing number of patients undergo computerized tomography perfusion (CTP) acutely, including some CVT patients. We aim to describe the benefit of adding CTP in CVT diagnosis compared to brain non-contrast CT (NCCT) and/or CT angiography (CTA). We estimated the sensitivity, specificity, predictive values and area under the curve (AUC) for each imaging modality.

Methods
Retrospective CVT patients presenting at a single comprehensive stroke centre (2001 -2021) who underwent acute CTP were included. CVT patients were analysed at a 1:5 ratio with patients presenting with stroke-like symptoms who underwent CTP. Imaging was reviewed by three neurologists blinded to diagnosis but unblinded to clinical presentation. Agreement by all assessors was required for an imaging finding to be assigned as positive. Receiver operating characteristic curve analysis was performed to estimate sensitivity, specificity, and area under the curve (AUC) for CVT detection.

Results
A total of 42 patients (7 with CVT and 35 controls) were analysed. The sensitivity of brain NCCT for CVT diagnosis was low (28.6%), with a high specificity (100%). The positive and negative predictive values were 100% and 87.5%, respectively. The discrimination of brain NCCT for CVT was moderate, with an AUC of 64.3 (95% CI: 46.2-82.4). Addition of CTA findings made no difference to the sensitivity or AUC. CTP review enhanced the sensitivity to 42.9%, and AUC to 71.4 (95% CI: 51.6-91.2).

Conclusion
CTP analysis in the hyperacute phase may provide opportunities to improve CVT detection.