Distinct long-term neurological trajectories and their predictors in cerebral venous sinus thrombosis: a latent class mixed model analysis.

Background: The long-term neurological recovery of patients with cerebral venous sinus thrombosis (CVST) exhibits significant heterogeneity, and the mechanisms underlying these distinct trajectories remain poorly understood. Objective: This study aimed to identify distinct long-term trajectories of functional outcome in CVST patients and to investigate their baseline predictors. Methods: In a longitudinal cohort of 127 CVST patients with complete follow-up data, we employed latent class mixed models (LCMM) to analyze repeated measures of the modified Rankin Scale (mRS) over time. The optimal number of trajectory classes was determined using established information criteria. Multivariable logistic regression was then used to identify baseline factors independently associated with class membership. Results: A two-trajectory model best characterized the heterogeneity in long-term outcomes. Class 1 (the "Poor-Recovery" Class, Conclusion: CVST patients follow one of two distinct long-term functional trajectories, which are predominantly driven by age and the status of the venous collateral circulation. These findings provide a mechanism-based framework for prognosis and highlight the potential for early risk stratification, paving the way for more personalized patient management.