Prognostic Utility of Noninvasive Brain Monitoring in Moderate-to-Severe Cerebral Venous Thrombosis: A Prospective...

Cerebral venous thrombosis (CVT) is a major cause of stroke in young adults, but existing prognostic scores rely only on clinical and radiologic data and may not reflect brain function. We evaluated the use of noninvasive multimodal brain monitoring (MBM) in moderate-to-severe CVT and its added prognostic value over the Cerebral Venous Thrombosis-Grading Scale (CVT-GS).

In this prospective observational study, 53 patients with moderate-to-severe CVT admitted to a tertiary neurosciences center (September 2021 to March 2023) underwent bedside MBM within 24 hours of admission. Tools included transcranial Doppler (TCD) for flow velocities, pulsatility index (PI), and autoregulation (transient hyperemic response ratio [THRR]); ultrasound for optic nerve sheath diameter (ONSD); bispectral index (BIS); and regional cerebral oxygen saturation (rSO₂). Neurological outcome was assessed at 1 month using the modified Rankin Scale (mRS). Predictors were analyzed using correlation and logistic regression. ROC curves were compared with the DeLong test.

At 1 month, 27 patients (50.9%) had a poor outcome (mRS ≥3), including 12 deaths (22.6%). Raised ONSD, elevated PI, impaired autoregulation (THRR ≤1.02), and reduced BIS were significantly associated with poor outcome and mortality, while rSO₂ and most TCD velocities were not. Adding MBM to CVT-GS improved accuracy: for mortality: adding ONSD and PI increasedAUC from 0.74 to 0.91; for poor outcome: addingTHRR and BIS increasedAUC from 0.76 to 0.92 (both P <0.05).

Noninvasive MBM can be used in patients with CVT. Integrating noninvasive surrogates of ICP, autoregulation, and brain electrical activity with CVT-GS improves outcome prediction.