RSNA 2014 

Abstract Archives of the RSNA, 2014


SSM07-03

Is It Possible to Accurately Diagnose Cerebral Venous Sinus Thrombosis on Noncontrast CT?

Scientific Papers

Presented on December 3, 2014
Presented as part of SSM07: ISP: Emergency Radiology (Neurologic Emergencies)

Participants

Jian Guan MD, Presenter: Nothing to Disclose
Ling Lin, Abstract Co-Author: Nothing to Disclose
Ling Zhang MD, Abstract Co-Author: Nothing to Disclose
Guangqi Yang, Abstract Co-Author: Nothing to Disclose
Zhiyun Yang, Abstract Co-Author: Nothing to Disclose

PURPOSE

With its highly variable clinical presentation, the diagnosis of cerebral venous sinus thrombosis (CVST) challenging. Noncontrast CT is still the preferred imaging exam in most emergency departments. The aim is to investigate the possibility of accurate diagnosis for CVST on noncontrast CT.

METHOD AND MATERIALS

There were 35 cases with CVST (case group) and 40 cases without CVST (control group). All cases received noncontrast CT scan. The two groups were matched by age and gender. Two radiologist blindly reviewed all CT examinations and made desision together. Another experienced radiologist measure CT value of cerebral venous sinus and summary clinical history and symptoms. Independent samples t test was performed between two groups, and receiver operating curve was performed to get critical value of venous sinus for diagnosis. Attenuation of venous sinus (critical value) (A), relative history (SLE, nephropathy, pregnancy etc)(B), symptoms (headache and/or spasm) (C),hemorrhagic infarct and/or epicranium swelling on CT (D) were taken as four variables. We calculated sensitivity, specificity and reliability of each variable for diagnosis of CVST. The impression were compared via grading of the mentioned variables (1-3 points) and the sum of each case in both groups were analyzed (Chi-square test / Fisher exact test). 

RESULTS

Attenuation of venous sinus in case group was significantly higher than control group (P<0.001). The best critical value for diagnosis was 68HU (Youden index = 0.886). The best single indicator was variable A (specificity 93%), followed by variable D (specificity 77%), while the other two indicators were of no statistical significance in two groups, According to the results, variable A was considered as 3 points, variable D as 2, both variable B and C as 1. The score of B,C or D variable can be accumulated repeatly. The case with total score no less than 5 points is highly correlated with clinical diagnosis as CVST in following up.The sensitivity, specificity and reliability were 83%, 100% and 94%, respectively ( P<0.01).

CONCLUSION

Combined with relative history, symptoms and noncontrast CT findings, we can make an accurate diagnosis for CVST.

CLINICAL RELEVANCE/APPLICATION

For the patients who have relative history, especially with headache and/or spasm, noncontrast CT is a valuable exam to exclude CVST and make a decision for further imaging.

Cite This Abstract

Guan, J, Lin, L, Zhang, L, Yang, G, Yang, Z, Is It Possible to Accurately Diagnose Cerebral Venous Sinus Thrombosis on Noncontrast CT?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14004378.html