Predictors of early clinical progression of cerebral venous thrombosis: A multicenter retrospective clinical study

Abstract

Abstract Background: Anti-coagulation is the mainstay treatment for cerebral venous thrombosis (CVT). A subset of patients might deteriorate despite anti-coagulation. Studies exploring the predictive performance of major risk factors associated with clinical progression (CP) are insufficient in CVT. This study aimed to investigate the predictive value of cerebrospinal fluid/serum albumin ratio (CAR) and other clinical parameters for CP and to guide endovascular therapy in the future for CVT patients. Methods: A retrospective review was performed from four medical centers between October 2012 and April 2021. The primary outcome was CP. The multiple logistic regression modeling, and receiver operating characteristic (ROC) curves were applied to determine the predictive value for deterioration. Results: We enrolled 123 patients with CVT. 108 (87.8%) were treated with anti-coagulation treatment, 31(25.20%) of whom underwent endovascular treatment. 36(29.27%) were found to have deterioration. The CAR was related to CVT severity, occurrence of progression and modified Rankin Scale score at discharge. The most important risk factors of deterioration in univariate analysis, included decreased level of consciousness, papilledema, epileptic seizure, infection, location score (number of sinuses involved) , D-dimer, and CAR. The multivariate logistic regression analysis included decreased level of consciousness, papilledema, location score (number of sinuses involved), D-dimer, and CAR. This model had a predictive ability to identify deterioration of 91.7%(AUC), with a sensitivity of 79.1% and a specificity of 62.4% in the ROC curve analysis. which was the largest of all predictive value. Patients with a CVT score of ≥6 have at least 80% chance of deterioration. Conclusions: Decreased level of consciousness, papilledema, number of involved sinuses, D-dimer, as well as CAR are important influence factors in predicting deterioration after CVT. CVT patients with the above clinical features may benefit from early endovascular therapy.