Retrospective analysis of cerebral MR venography was performed with blinding to other MR sequences and clinical data to evaluate the number of patent cortical veins, the caliber of cortical veins, patency of dural sinuses as well as the status of all venous drainage pathways including bilateral ophthalmic veins, petrosal sinuses, calvarial veins, facial veins, and scalp veins. In the second session, cortical vein clot number and dural sinus clot burden were evaluated with blinding to MRV and clinical data.  Finally, cerebral hemispheres with and without infarction were identified independently. Statistical analysis was performed with Wilcoxon paired sample test and logistic regression models.

Among 40 cases of acute cerebral venous thrombosis, there were 14 males and 26 females in the age range of 3-70 years, (mean 32 ± 19 SD). Venous infarct was present in 30 of 80 cerebral hemispheres. Univariate analysis revealed significant association with venous infarct for the number of cortical vein clots, cortical vein patency and sinus clot burden (p < 0.05). On multivariate analysis, only the number of cortical vein clots revealed significant association with infarct (p < 0.0001). Among the venous channels for drainage, only the left sphenoparietal sinus and the left inferior petrosal sinus revealed significant association with venous infarcts (p < 0.05).

Majority of the venous drainage pathways has no influence whereas the number of cortical vein clots has significant association with the development of venous infarcts in acute cerebral venous thrombosis.