The Abnormalities Of Diffusion-Weighted MRI And Its Clinical Significance In Transient Ischemic Attacks

Purpose:The purpose of this study was to determine characteristics in the clinical background of TIA patients with DWI abnormal,to assessed the utility of DWI in patients with TIA.Methods:A retrospective analysis was performed on all patients entered in our hospital during 2006.4-2007.6 who were clinically diagnosed with a TIA and who had also undergone a DWI scan＜5 days after symptom onset.Record clinical data and results of every assisted examination.Deduce underlying pathology,anatomic and vascular localization of TIAs with and without imformations provided by DWI.We examined the frequency of TIA-related DWI abnormalities.The relationship between stroke classfication and asymptomatic lesions were then analyzed.We compared the clinical background of TIA patients with symptomatic lesions,patients with asymptomatic lesions,and patients without DWI abnormalities respectively.The relationship between DWI lesion patterns and underlying pathology were analyzed. We compared initial diagnoses of underlying pathology,anatomic and vascular localization between final diagnoses with imformations provided by DWI.We compared the prognosis of TIA patients with or without DWI abnormalities. Statistical analysis were performed with the SPSS statistical package(11.5)。Results:(1)78 patients with TIA(34.21%)had DWI abnormalities,whereas 150 (65.79%)had no abnormalities.71(31.14%)patients with TIA had symptomatic lesion.7(8.77%)patients only had asymptomatic lesions.Asymptomatic lesions was associated with CE and≥2 etiologies.(2)TIA duration≥30 minutes,aphasia, paresis,atrial fibrillation,identified stroke mechanism,the badness of collateral vessels were observed more frequently in the symptomatic lesions positive group than in the negative group.No difference in other symptoms,vascular risk factors,history of either stroke or TIA,history of ischemic heart disease,delay from onset to DWI,number of TIAs before DWI and so on.(3)Atrial fibrillation was observed more frequently in the asymptomatic lesions group than in the negative group.There were no other diffenence between thease two groups.(4)40 patients had single lesions,23 patients had multiple lesions in one vascular territory,15 patients had multiple lesions in multiple vascular territories.A single corticosubcortical lesion was associated with LAA,compared with other DWI lesion patterns.Multiple lesionin multiple cerebral circulations was associated with CE.subcortical lesion(＜15mm)was associated with SVO.(5)Diffusion MRI information changed the suspected etiologic mechanism in 12(15.83%)patients with DWI abnormalities,changed the suspected anatomic and vascular localization in 20(25.64%)patients with DWI abnormalities.(6)In the first few weeks after a TIA,positive group were more likely to have recurrent stroke than negative group,TIA patients with multiple lesions were more likely to have recurrent stroke than patient with single lesionsConclusions:(1)More than 1/3 patients with TIA had DWI abnormalities,some of their lesion were symptomatic lesions.Asymptomatic lesions provided some clue for stroke etiologic subtypes.(2)TIA patients with symptomatic lesions had some special clinical characteristics in the clinical background,such as had long duration time,had identified stroke mechanism and so on.(3)TIA patients with asymptomatic lesions had no special clinical characteristics.(4)Different topographic patterns in TIA patients with DWI abnormalities may be related to underlying pathology.(5) Diffusion MRI information changed the suspected etiologic mechanism and the suspected anatomic and vascular TIA localization in patients with DWI abnormalities. (6)Imformations from DWI were useful in predicting recurrent stroke.