Silent Diffusion-Weighted Imaging-Positive Lesions In Acute Ischemic Strokes:Prevalence,Etiology And Prognosis

Background and purpose:With the popularization of magnetic resonance imaging(MRI)technology and the aging of social population,silent diffusion weighted imaging(DWI)lesions can be observed occasionally in an increasing number of people on brain MRI.The main causes of silent DWI lesions include cerebral small vessel disease(CSVD),embolism and hypoperfusion,and may lead to an increased risk of future stroke.When silent DWI lesions are revealed outside the vascular territory of acute infarcts in acute ischemic stroke patients,it is important to determine the underlying causes and clinical relevance for guiding secondary stroke prevention therapies.To date,there is a lack of studies on silent DWI lesions outside the vascular territory of acute infarct lesions in acute ischemic stroke patients.In this study,we sought to systematically investigate the prevalence,etiology and clinical relevance of silent DWI lesions outside the vascular territory of acute infarct lesions in acute ischemic stroke patients.Methods:This study is based on a national multicenter,prospective cohort study named stroke imaging package study(SIPS).The inclusion criteria are:1)18-80 years old,the acute first-ever ischemic stroke,and the time from onset to MRI examination is within 72 hours;2)the patient’s vital signs were stable;3)patients receiving thrombolytic/intravascular therapy can participate in the study.Silent DWI lesions and their imaging characteristic were determined by DWI and ADC maps.Extracranial or intracranial large artery stenosis outside the vascular territory of acute infarct lesions were examined by traditional vascular lumen imaging examinations(e.g.magnetic resonance angiography,MRA)and heart diseases(such as atrial fibrillation or flutter)were determined by electrocardiogram and echo cardiography.The presence of intracranial nonstenotic plaque outside the vascular territory of acute infarct lesions were examined by three-dimensional T1 weighted high-resolution magnetic resonance imaging(3D T1 HR-MRI)and CSVD imaging markers were evaluated by conventional brain MRI sequences.In addition,we also collected detailed baseline data of patients,including demographic data,traditional vascular risk factors,National Institution of Health Stroke Scale(NIHSS)score at admission and related treatments.The primary endpoints were defined as a poor 90-day functional outcome(modified Rankin Scale score>2)and stroke recurrence.We further analyzed the prevalence of silent DWI lesions outside the vascular territory of acute ischemic strokes,and it’s correlation with embolic sources(including artery-to-artery embolism or cardioembolism),intracranial nonstenotic plaques,CSVD,and outcomes in acute ischemic stroke patients.Results:A total of 756 patients(538 men[71.2%];age[mean±SD]:60.5±11.1 years)were included for analysis.Forty-one silent DWI lesions were detected in 34(4.5%)of the enrolled 756 patients.The lesions were usually small(<20 mm,41/41,100%),single(29/34,85.3%),and seldom involving cortical regions(4/34,11.8%).There was no significant difference between patients with and without silent DWI lesions in embolism sources(29.4%vs 28.8%;p=0.94)or intracranial nonstenotic plaques(35.3%vs 41.4%;P=0.48)outside the vascular territory of acute infarct lesions.Compared with patients without silent DWI lesions,patients with silent DWI lesions had higher CSVD burden,including higher total CSVD score(3[2～3]vs 1[0～2];p<0.001),and higher proportion of severe white matter hyperintensities(WMH)(58.8%vs 25.4%;p<0.001),lacunes(61.8%vs 36.0%;p=0.002)and moderate-severe basal ganglia enlarged perivascular spaces(BGPVS)(65.6%vs.34.9%;p<0.001).Multivariate regression analysis showed that not embolism sources(odds ratio[OR]=0.782,95%confidence interval[CI]=0.283-1.873,p=0.51)but severe WMH(OR=3.026,95%CI=1.249-7.3325,p=0.01),moderate-severe BGPVS(OR=2.458,95%CI=1.0605.699,p=0.04),lacunes(OR=2.387,95%CI=1.021-5.584,p=0.04)and statin treatment(OR=0.297,95%CI=0.112-0.783,p=0.01)were independent predictors of silent DWI lesions.Silent DWI lesions were not associated with poor functional outcome(OR=1.610,95%CI=0.561-4.623,p=0.38)and stroke recurrence(OR=3.416,95%CI=0.654-14.082,p=0.15).Conclusions:Silent DWI lesions are not uncommon in ischemic stroke patients and are most likely caused by coexistent active CSVD.Our results do not support aggressive antiplatelet or anticoagulation therapy in acute ischemic stroke patients with silent DWI lesions outside the vascular territory of acute infarct lesions.