| Literature DB >> 29921486 |
Xiaodi Li1, Yuzhou Wang2, Zhanhang Wang1, Jianjun Lu1, Yan Xu1, Jinlong Ye1, Zuying Kuang1, Bo Li1, Mengqiu Pan1, Wenming Chen1, Shuisheng Lu3, Ziyang Zhou3, Lina Cheng3, Honghao Wang4.
Abstract
To compare the utility and limitation of magnetic resonance spectroscopy (MRS) and arterial spin labeling (ASL) in the differentiation between mitochondrial encephalomyopathy, lactic acidosis, plus stroke-like episodes (MELAS) and acute ischemic stroke (AIS), a retrospective review of 17 MELAS and 26 AIS patients were performed. In all patients both MRS and ASL scans were performed within 1 week after admission. Demographic, clinical, laboratory and MR imaging data were reviewed and compared between the two groups. Compared with AIS, MELAS patients had a younger age of onset, a longer disease duration, a higher occurrence of epilepsy attack, occipital and parietal lesions, and dilated cerebral arteries (P < 0.05). In all MELAS patients lactate peak and hyperperfusion of the lesion was revealed. However in AIS lactate peak was observed in only 69.2% and hyperperfusion was observed in only 34.6% ischemic lesions (P < 0.05). Choline/Creatine ratios and Lactate/Creatine ratios were higher in AIS, while in MELAS cerebral blood flow and lesion-normal perfusion ratio was much higher (P < 0.05). No correlations was found between metabolite ratios and perfusion parameters in either group (P > 0.05). Area under curve (AUC) of perfusion for the differentiation between MELAS and AIS was 0.958 (P < 0.001). The cut-off value was 2.075, with a sensitivity of 88.2% and a specificity of 96.2%. AUC of Lactate/Creatine ratio was 0.469 (P = 0.737). Utility of MRS is limited in the differentiation between MELAS and AIS, while MR perfusion profiles are much more sensitive and specific.Entities:
Keywords: Arterial spin labeling; Ischemic stroke; Lactic acidosis, and stroke-like episodes; Magnetic resonance spectroscopy; Mitochondrial encephalomyopathy
Mesh:
Year: 2018 PMID: 29921486 DOI: 10.1016/j.jocn.2018.06.015
Source DB: PubMed Journal: J Clin Neurosci ISSN: 0967-5868 Impact factor: 1.961