Researchers have developed the DIAS3 score, a six-factor tool that predicts the risk of epilepsy after cerebral venous thrombosis, estimating patient risk from 7% to 83% over three years, according to a recent study.
The DIAS3 score, a clinical tool developed to assess epilepsy risk following cerebral venous thrombosis (CVT), stratifies patient risk profiles based on six readily available clinical variables chosen from existing literature and clinical plausibility. This cohort study, published in JAMA Neurology, spanned multiple countries from 1994 to 2022 and included 2,937 patients from three registries, with 1,128 used for derivation and the ACTION-CVT (n = 543) and Israel CVT (n = 556) cohorts used for external validation.
In the derivation cohort, 11% of patients developed post-CVT epilepsy over a median follow-up of 12 months, with predicted risks ranging from 7% to 68% at one year and up to 83% at three years based on individual factors. Key predictors in the DIAS3 score included age, decompressive hemicraniectomy, intracerebral hemorrhage, acute-phase seizures, status epilepticus, and subdural hematoma. Model discrimination was strong, with C statistics of 0.74 (95% confidence interval [CI], 0.70-0.79) in derivation, 0.76 (95% CI, 0.67-0.84) in ACTION-CVT, and 0.80 (95% CI, 0.75-0.86) in Israel CVT for one-year outcomes. Calibration indicated good agreement, with slight overestimation in low-risk and underestimation in high-risk patients.
The DIAS3 score may help guide antiseizure medication (ASM) decisions, identifying patients who may not require prolonged ASM treatment. With a median time to the first late seizure of five months post-CVT, the score provides early seizure risk assessment. Limitations include retrospective data components and absent three-year validation for Israel CVT, affecting the model's accuracy for longer-term outcomes. Further research is needed to evaluate the impact of ASM timing, duration, and type on epilepsy risk post-CVT. The DIAS3 calculator is accessible online and can be used across clinical settings for individualized risk assessment in CVT care.
Full disclosures are available in the published study.