Abstract

Background The stage of cerebral venous thrombosis (CVT) is crucial to guide treatment decisions. This study aims to examine changes in fibrinolytic indicators throughout CVT onset and validate a predictive model using admission fibrinolytic indicators to estimate the CVT stage. Methods Retrospective analysis was conducted on data from 292 CVT patients. We utilized linear regression, time series, and univariate ANOVA analyses to explore characteristics of change in fibrinolytic indicators with CVT duration and identified time point at which fibrinolysis indexes showed significant changes as the time point for acute and chronic stages of CVT. A nomogram was employed to construct a prediction model using a training set, which was then evaluated for discrimination, calibration, and clinical utility. Results Prolonged onset duration independently correlated with decreased fibrinogen and D-dimer after adjusting for all variables, with adjusted correlation coefficients of −0.003 (−0.005, −0.001) and −0.004 (−0.007, −0.001), respectively. Significant changes in fibrinolytic indicators were observed around 14 days after CVT onset. The training set demonstrated an area under the curve (AUC) of 0.851 (95% CI: 0.7989-0.904) for the prediction model. Internal validation showed that the nomogram accurately predicted acute CVT with an AUC of 0.828 (95% CI: 0.738-0.918). Conclusion According to the trend of fibrinolysis index, 14 days of onset can be used as the dividing point of acute and chronic stages of CVT. For patients with unclear onset, the present model, based on admission fibrinogen and D-dimer values, can accurately predict the stage of CVT. The high discriminative ability indicates the potential of this model for classifying the acute patient.