Cerebral Venous Thrombosis Prediction Score in a Real-World Setting: A Retrospective Single-Centre Cohort Study of Patients with Confirmed Diagnosis

Background/Objectives: In 2020, a clinical score to predict the risk of cerebral vein thrombosis was published. The score classifies patients into risk categories: low (<3 points), moderate (3–5 points) and high (>5 points). The low-risk result of the score combined with D-dimer level could potentially exclude cerebral vein thrombosis. The main aim of our study was to assess utility of the score in a group of patients with confirmed cerebral vein thrombosis in a real-world setting. Methods: A retrospective assessment of patients with a confirmed diagnosis of cerebral vein thrombosis admitted to a department of neurology was performed. Data on patients’ demographics and medical history, including the components of the score, were collected. Based on the inclusion criteria from the original publication, the number of patients eligible and ineligible for using the score was calculated. Next, the patients were assessed with the score. Results: Thirty-seven patients with newly diagnosed cerebral vein thrombosis were included. Twenty-three (62%) were eligible according to the inclusion criteria of the original study. The sensitivity of the score alone for this group was 52% at a cutoff of three points, and in combination with D-dimer level it was 92%. Two patients were classified as low-risk according to the score and had low D-dimer level. In the remaining symptomatic patients, the score had a sensitivity of 44% without D-dimer testing, and 89% with D-dimer. D-dimer level alone had a sensitivity of 87% and 67%, respectively. Conclusions: The score in combination with D-dimer level had higher sensitivity than D-dimer testing alone. However, the score combined with D-dimer level could not exclude cerebral vein thrombosis in the low-risk patients.