Application Of Multimodal MR Imaging In Early Diagnosis And Prognostic Evaluation Of Cerebral Venous Sinus Thrombosis

Objective:Cerebral venous and sinus thrombosis（CVST）is a relatively rare clinical disease of the intracranial venous system.Although the incidence is low,due to the complex etiology and lack of specific clinical symptoms,it is easy to missed diagnois and misdiagnosis,leading to poor outcomes and high mortality for patients who failed to be diagnosed and treated early.Recently,the widespread application of imaging technology,especially magnetic resonance imaging（MRI）,has increased the diagnosis rate of CVST,However,its early diagnosis accuracy is not high,and predictors of outcome ave rarely evaluated in properly powered studies.The purpose of this research was to investigate the etiology,clinical symptoms,imaging manifestations and prognosis,and to analyze the prognostic factors related to CVST,and to guide clinical precision treatment and prognostic evaluation.Methods:This paper retrospectively analyzed the etiology,clinical manifestations,laboratory examinations,conventional MRI scans,magnetic resonance venography（MRV）,susceptibility weighted imaging（SWI）,diffusion weighted imaging（DWI）and other multimodal MRI data and prognosis of 70 CVST patients admitted to the First Hospital of Shanxi Medical University from January 2011 to December 2020.To evaluate the value of conventional MRI scans,MRV and SWI alone and combination in diagnosing CVST,including accuracies,sensitivities,specificities,positive predictive values and negative predictive values.The etiology,clinical symptoms,imaging manifestations and other factors of prognosis were compared.Results:Among 70 CVST patients,36 were females,34 were males,and the ratio of male to female was 1:1.1,the average age is 38.41±16.46 years old.Infection was the most common risk factor,accounting for 18.57%.Headache was the most common clinical symptom,accounting for 77.14%.The transverse sinus thrombosis was the most common in patients（70%）,followed by the superior sagittal sinus（65.71%）.64.29%of patients had secondary brain parenchymal lesions.The differences in the detection of cerebral venous sinus thrombosis,draining venules,circuitous drainage veins and microhemorrhage between conventional MRI scans,MRV and SWI were statistically significant.Among them,MRV was the most sensitive techniques in the detection of CVST and draining venules.SWI was the most sensitive techniques in the detection of circuitous drainage veins and microhemorrhage,and the sensitivity and accuracy of the combination of three MR sequences was the highest.The Receiver operating characteristic curve（ROC）displayed that the area under curve（AUC）of conventional MRI scans,MRV,SWI and combination were 0.630,0.780,0.633 and 0.888respectively.AUC_combination was significantly higher than AUC_{conventional MRI scans},AUC_MRV and AUC_SWI,and the differences were statistically significant（Z values were 3.632,2.321 and 3.567,p<0.05）.Univariate analysis showed that the differences in infection,seizure,paresis,mental status disorder,brain parenchymal lesions,brain edema,hemorrhage,intracranial infarction,collateral formation,and grade III collateral formation between the good prognosis group and the poor prognosis group were statistically significant.Multivariate Logistic regression analysis showed that dyskinesia and brain parenchymal lesions were independent predictors of poor prognosis（OR=0.196,95%CI:0.046-0.828,p=0.027;OR=0.028,95%CI:0.002-0.521,p=0.016）,and collateral formation was a protective factor for the patient’s prognosis（OR=7.262,95%CI:1.170-45.063,p=0.033）.Conclusion:MRV was the most sensitive techniques in the detection of cerebral venous sinus thrombosis and draining venules.SWI was the most sensitive techniques in the detection of circuitous drainage veins and microhemorrhage,and the sensitivity and accuracy of the combination of three MR sequences was the highest.Paresis and secondary brain parenchymal lesions were independent risk factors for poor prognosis of CVST,and collateral formation was an independent predictor of good prognosis.