The Application Of Asymmetric Hypointensive Cerebral Veins On Susceptibility-weighted Imaging In Intraveous Thrombolysis For Acute Ischemic Stroke

Intravenous thrombolysis with recombinant tissue plasminogen activator (rtPA) is the guideline-recommended therapy for hyperacute ischemic stroke. The purpose of thrombolysis is to salvage the ischemic penumbra, defined as ischemic but still viable tissue which will die if reperfusion is not established timely.Positron emission tomography (PET) is currently the golden standard to evaluate ischemic penumbra in human. However, its application is limited due to the long time of scan and post-processing. With the advance of modern imaging, magnetic resonance imaging (MRI) is increasingly used in the clinical practice. The perfusion-diffusion mismatch model, based on a mismatch between the lesion volumes on perfusion-weighted imaging (PWI) and diffusion-weighted imaging (DWI), has been proposed as a method to select patients for reperfusion therapy. Despite theoretical evidence, there is no consensus about the optimal PWI parameter to accurately define the ischemic tissue and it is uncertain whether PWI lesion volumes overestimate the penumbra tissue.More recently, susceptibility-weighted imaging (SWI), which originally called blood oxygen level dependent (BOLD) venographic imaging, has demonstrated advantages in the detection paramagnetic substances such as deoxyhemoglobin. The intravascular deoxygenation leads to a signal drop along the course of cerebral veins on SWI venography. Thus SWI venographymay potentially reflect the hypoxia status. Based on this assumption, we aimed to investigate the clinical use of hypointensive cerebral veins on SWI and the possible underlying mechanisms. Part one:Asymmetric hypointensive cerebral veins on SWI correlates with the volume of hypoperfusionPurpose:We hypothesize that the asymmetric hypointensive signal of cerebral veins on SWI correlates with the volume of hypoperfusion on perfusion-weighed imaging in acute ischemic stroke patients within onset of6hours.Methods:We conducted a retrospective study of anterior circulation ischemic stroke patients with SWI/DWI acquired before thrombolysis. Tmax>6s on PWI was defined as hypofusion lesion. Asymmetry index (AI) was defined as the ratio of cerebral vein voxel count between the ischemic and normal hemisphere on the SWI phase map. Absolute△AI was defined as the difference between baseline AI and24hours AI. Relative△AI was defined as the ratio between baseline AI and24hours AI. We used SWI-ASPECTS and AI of cerebral veins respectively to test whether asymmetric hypointensive signal on SWI was associated with the volume of ischemic penumbra. We then compared the AAI between patients with or without recanalization/reperfusion.Results:Patients with low SWI-ASPECTS had lower TTMI grading score (1±1vs2±1, P<0.001), larger hypoperfusion volume (105±62vs57±89ml, P=0.015) and larger hypoperfusion lesion volume (84±53vs35±52ml, P=0.001), compared with high SWI-ASPECTS group. Patients with high AI score had lower TTMI grading score (1(IQR:0-1) versus2(IQR:0-3), P=0.002) and larger PWI lesion volume (120±83ml versus58±67ml, P=0.004) than those with low AI score. Both absolute and relative△AI were higher in patients with recanalization (absolute AAI:1.20versus0.11, P<0.001; relative△AI:3.54versus1.11, P<0.001) and reperfusion (absolute AAI:0.93versus0.12, P=0.031; relative AAI:2.06versus1.17, P=0.029) than those without.Conclusion:Asymmetric hypointensive signal of cerebral veins on SWI, which were reversed after successful recanalization, can reflect the hypoperfusion lesion volume in acute ischemic stroke patients within6hours. Part two:Susceptibility-diffusion mismatch predicts thrombolytic outcomes:a retrospective studyBackground and Purpose:Asymmetric hypointensity of cerebral veins on susceptibility-weighted imaging (SWI) has been shown to indirectly reflect tissue hypoxia after cerebral ischemia. We therefore investigated whether patients with prominent asymmetry of cerebral veins on SWI and a relatively small diffusion-weighted imaging (DWI) lesion (SWI-DWI mismatch), representing the presence of salvageable tissue, were more likely to benefit from thrombolytic therapy.Methods:We conducted a retrospective study of anterior circulation ischemic stroke patients with SWI/DWI acquired before thrombolysis. Asymmetry index (AI) was defined as the ratio of cerebral vein voxel count between the ischemic and normal hemisphere on the SWI phase map. We defined SWI-DWI mismatch as an AI score>1.75with a DWI lesion volume≤25ml. Favorable outcome was defined as modified Rankin Scale (mRS)0-2at3months. Univariate and multivariate logistic-regression analyses were used to examine the association between mismatch profile and favorable outcome.Results:Fifty-four thrombolytic patients were enrolled in this study. Patients with SWI-DWI mismatch achieved higher rate of favorable outcome, compared with those without (78%versus44%; adjusted odds ratio=6.317,95%CI:1.12-35.80, P=0.037). Patients with SWI-DWI mismatch were also more likely to obtain favorable outcome from reperfusion (91%versus43%, P=0.033) or recanalization (100%versus40%, P=0.013). The accuracy of SWI-DWI mismatch for predicting favorable outcome was higher than perfusion-diffusion mismatch (63%versus48%).Conclusion:The presence of SWI-DWI mismatch may identify ischernic patients who will benefit from early reperfusion therapy.