Abstract
241254
Introduction: Symptomatic patients with intracranial arterial stenosis or occlusion are at extremely high risk of recurrent ischemic attacks or cerebral infarction, who should be performed with aggressive intervention. Therefore, it’s essential to perform accurate evaluations of their cerebrovascular damage as indications for surgical treatment. Cerebral vascular reserve (CVR) is an independent hemodynamic risk factor that predicts long-term poor prognosis in patients with intracranial artery stenosis. Recent studies have shown that resting-state blood oxygenation level-dependent (BOLD) obtained by MRI may be a valuable alternative in detecting CVR in clinical applications when gas challenge is not feasible. On the other hand, 18F-FDG PET reflects the metabolic deficit of ischemic brain tissue, which is closely related to the patient's neurological dysfunction. However, reliable combined imaging markers have not been established to evaluate CVR and cerebral metabolic damage in a single scanning process. This study aims to simultaneously evaluate the CVR and brain metabolism of high-risk patients with atherosclerotic intracranial artery stenosis using 18F-FDG PET/MR and further explore their value on the follow-up after surgical treatment.
Methods: Forty-five symptomatic patients with severe unilateral middle cerebral artery (MCA) or internal carotid artery (ICA) stenosis or occlusion and 20 age- and sex-matched healthy controls were included, while 16 of the patients were followed up after extracranial and intracranial (EC-IC) bypass surgery. 18F-FDG PET and resting-state BOLD MRI were collected by a hybrid PET/MR (Signa, GE Healthcare, USA). National Institutes of Health Stroke Scale (NIHSS), and modified Rankin Scale (mRS) were collected to evaluate their neurological performance. The voxel-wise values of resting-state BOLD-CVR and PET-SUVR were calculated and extracted from bilateral MCA territories using Matlab R2019b (MathWorks, USA). The asymmetry index (AI) was used to evaluate the interhemispheric asymmetry of CVR and SUVR. The predictive values of SUVR-AI and CVR-AI in the detection of patients were analyzed by the receiver operating characteristic curve (ROC). The CVR/SUVR grading criteria (Grade I, CVR normal, SUVR normal; Grade Ⅱ, CVR decreased, SUVR normal; Grade Ⅲ, CVR normal, SUVR decreased; Grade Ⅳ, CVR decreased, SUVR decreased) was established to distinguish the severity of hemodynamic and metabolic injury. Group differences of CVR-AI, SUVR-AI, and CVR/SUVR grades were compared between the pre- and post-operative scans.
Results: Visual assessment showed 95.6% and 57.8% of 45 pre-operative patients appeared CVR and SUVR reduction respectively. The threshold of CVR-AI with the best predictive efficiency in discriminating patients from controls was 10.25% (AUC=0.762), while the threshold of SUVR-AI in discriminating patients with neurological disorder or not was 11.67% and 11.10% (AUC=0.950 and 0.874). Combining SUVR-AI and CVR-AI had the highest predictive efficiency in discriminating patients from controls (AUC=0.848). According to CVR/SUVR grading criteria (with the threshold of AI=10%), significant differences between groups were detected in MCA-CVR, CVR-AI, MCA-SUVR, SUVR-AI, infarct volume, NHISS and mRS (P<0.05). For the 16 patients who underwent EC-IC treatment, postoperative CVR-AI, SUVR-AI, and CVR/SUVR grades significantly decreased than those before surgery (P<0.05, Fig.1 and Fig.2A). A significant positive correlation was found between the preoperative CVR/SUVR grades and the variance of the grades after surgery (r= 0.578, Fig.2B).
Conclusions: The combined use of 18F-FDG PET and resting-state BOLD by using PET/MR could provide a clinically feasible criterion for the simultaneous evaluation of CVR and metabolic damage, and thus contribute to the risk stratification of patients with intracranial artery stenosis before surgery.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.