TY - JOUR
T1 - Significance of hypointense leptomeningeal vessels on T2*-weighted gradient echo imaging for prediction of acute ischemic lesion growth
AU - Cho, Kyung Hee
AU - Kim, Hye J.
AU - Cho, A. Hyun
AU - Kwon, Sun U.
AU - Kim, Jong S.
AU - Kang, Dong Wha
N1 - Copyright:
Copyright 2008 Elsevier B.V., All rights reserved.
PY - 2007/11/13
Y1 - 2007/11/13
N2 - Background and Purpose: For the response to reduced cerebral perfusion pressure, oxygen extraction fraction (OEF) increases in brain regions with reduced cerebral blood flow to preserve oxygen metabolism. T2*-weighted gradient echo imaging (GRE) is sensitive to blood deoxygenation, which results in hypointense signals in blood vessels. Thus, hypointense leptomeningeal vessels (HLV) on GRE in hypoperfused region may indicate the presence of tissue with increased OEF. The relationship of this vessel sign with regional tissue fate remains unknown in acute ischemic stroke. We sought to examine the prognostic value of the HLV on GRE for the prediction of acute ischemic lesion growth. Methods: We included consecutive acute ischemic stroke patients who 1) had an acute infarction in the territory of middle cerebral artery (MCA), 2) underwent diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI) and GRE within 24 hours of stroke onset and follow-up DWI 5 days later, and 3) had a baseline PWI-DWI mismatch (mean transit time lesion greater than DWI lesion >20%). Patients receiving thrombolysis or neurointerventional therapy were excluded. The presence of HLV on GRE within the territory of mean transit time deficit was determined. Patients were divided into 3 groups: Group A, no HLV; Group B, HLV with increased cerebral blood volume (CBV); and Group C, HLV without increased CBV. The ischemic lesion volume was measured blind to the clinical and other imaging information. The infarct volume on baseline and follow-up DWIs and the difference between the two DWIs were compared among 3 groups. The tissue fate of the area with HLV was determined blind to the information of CBV. Results: Of the 81 patients who met the inclusion criteria, HLV was observed in 50 (61.7%). CBV was increased in 34 of these 50 patients (68.0%). The ischemic lesion growth (18.4 cc in Group A (n=31), 23.2 cc in Group B (n=34), and 28.7 cc in Group C (n=16)) was significantly different among 3 groups (p=0.030, Kruskal-Wallis test), with infarct growth being the greatest in Group C: there was no difference between Groups A and B (p=0.46), while significant difference existed between Groups A and C (p=0.020) and between Groups B and C (p=0.017). In terms of tissue fate of the area with HLV, progression to infarction was more frequent (p=0.005) in Group C (9 of 16, 56.3%) compared to Group B (5 of 34, 14.7%). Conclusions: The presence of HLV on GRE, when associated with no increase of CBV, may indicate a hemodynamic marker of impending irreversible ischemic injury.
AB - Background and Purpose: For the response to reduced cerebral perfusion pressure, oxygen extraction fraction (OEF) increases in brain regions with reduced cerebral blood flow to preserve oxygen metabolism. T2*-weighted gradient echo imaging (GRE) is sensitive to blood deoxygenation, which results in hypointense signals in blood vessels. Thus, hypointense leptomeningeal vessels (HLV) on GRE in hypoperfused region may indicate the presence of tissue with increased OEF. The relationship of this vessel sign with regional tissue fate remains unknown in acute ischemic stroke. We sought to examine the prognostic value of the HLV on GRE for the prediction of acute ischemic lesion growth. Methods: We included consecutive acute ischemic stroke patients who 1) had an acute infarction in the territory of middle cerebral artery (MCA), 2) underwent diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI) and GRE within 24 hours of stroke onset and follow-up DWI 5 days later, and 3) had a baseline PWI-DWI mismatch (mean transit time lesion greater than DWI lesion >20%). Patients receiving thrombolysis or neurointerventional therapy were excluded. The presence of HLV on GRE within the territory of mean transit time deficit was determined. Patients were divided into 3 groups: Group A, no HLV; Group B, HLV with increased cerebral blood volume (CBV); and Group C, HLV without increased CBV. The ischemic lesion volume was measured blind to the clinical and other imaging information. The infarct volume on baseline and follow-up DWIs and the difference between the two DWIs were compared among 3 groups. The tissue fate of the area with HLV was determined blind to the information of CBV. Results: Of the 81 patients who met the inclusion criteria, HLV was observed in 50 (61.7%). CBV was increased in 34 of these 50 patients (68.0%). The ischemic lesion growth (18.4 cc in Group A (n=31), 23.2 cc in Group B (n=34), and 28.7 cc in Group C (n=16)) was significantly different among 3 groups (p=0.030, Kruskal-Wallis test), with infarct growth being the greatest in Group C: there was no difference between Groups A and B (p=0.46), while significant difference existed between Groups A and C (p=0.020) and between Groups B and C (p=0.017). In terms of tissue fate of the area with HLV, progression to infarction was more frequent (p=0.005) in Group C (9 of 16, 56.3%) compared to Group B (5 of 34, 14.7%). Conclusions: The presence of HLV on GRE, when associated with no increase of CBV, may indicate a hemodynamic marker of impending irreversible ischemic injury.
UR - http://www.scopus.com/inward/record.url?scp=36348973961&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=36348973961&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:36348973961
SN - 0271-678X
VL - 27
SP - BP15-02U
JO - Journal of Cerebral Blood Flow and Metabolism
JF - Journal of Cerebral Blood Flow and Metabolism
IS - SUPPL. 1
ER -