髓鞘少突胶质细胞糖蛋白抗体相关脱髓鞘病与水通道蛋白4抗体阳性视神经脊髓炎谱系疾病的磁共振特点分析

Magnetic resonance imaging characteristics of brain lesions in myelin oligodendrocyte glycoprotein antibody associated demyelinating diseases and aquaporin-4 antibody positive neuromyelitis optica spectrum disorders

目的探讨髓鞘少突胶质细胞糖蛋白(MOG)抗体相关脱髓鞘病与水通道蛋白4(AQP4)抗体阳性视神经脊髓炎谱系疾病(NMOSD)患者头颅磁共振成像(MRI)的病变分布、形态学特征及其在早期诊断中的临床价值。方法回顾性收集郴州市第一人民医院和青岛大学附属医院2018年9月至2021年6月经临床及血清抗体检测确诊的MOG抗体相关脱髓鞘病患者35例[男性20例,女性15例,年龄31(25,43)岁]和AQP4抗体阳性NMOSD患者36例[男性3例,女性33例,年龄42(29,54)岁],分别纳入MOG组和AQP4阳性组。所有患者均在治疗前行常规头颅MRI扫描,并记录颅内病变的部位、形态及数量。采用Wilcoxon秩和检验比较两组间不同类型的病灶数量,采用Logistic回归分析评价不同病灶对于两种疾病的提示意义。结果在22种不同类型的病灶中,共有7种类型的病灶数量组间差异有统计学意义,Logistic回归分析结果显示皮质及皮髓质交界病变(OR=21.91,95%CI 3.09~61.69,P<0.05)和幕下周围型白质病变(OR=10.48,95%CI 2.00~18.89,P<0.05)是鉴别MOG组和NMOSD组患者最重要的危险因素。皮质及皮髓质交界病变在MOG组中的发生率为51.4%(18/35),显著高于AQP4阳性组(2.8%,1/36),差异具有统计学意义(χ2=19.02,P<0.01)。幕下周围型白质病变在MOG组中的发生率为31.4%(11/35),显著高于AQP4阳性组(5.6%,2/36),差异具有统计学意义(χ2=6.31,P<0.05)。受试者工作特征曲线显示周围型病变(包括幕上软脑膜炎、皮质脑炎、皮质及皮髓质交界病变、幕下软脑膜炎、幕下软脑膜下脱髓鞘、幕下周围型病变6类病灶)对于提示MOG抗体相关脱髓鞘病颅内病变[受试者工作特征曲线下面积(AUC)=0.93]较仅依据皮质及皮髓质交界病变(AUC=0.75)或中央型病变(幕上脑室旁白质病变、间脑、幕下脑室旁等3类病变,AUC=0.64)具有更高的诊断效能。结论 MOG抗体相关脱髓鞘病较AQP4阳性视神经脊髓炎谱系疾病颅内病灶的发生率更高,且两组疾病的颅内MRI病灶在分布及形态上有特征性表现。识别周围型病灶(沿软脑膜分布)和中央型病灶(沿室管膜分布)分布模式对于鉴别两组疾病有一定参考意义。

Objective To investigate the distribution and morphological characteristics of brain magnetic resonance imaging(MRI)lesions in patients with myelin oligodendrocyte glycoprotein(MOG)antibody related demyelinating diseases and aquaporin-4(AQP4)antibody positive neuromyelitis optica spectrum disorders(NMOSD)and their clinical value in early diagnosis.Methods A total of 35 patients with MOG antibody related demyelinating diseases[20 males and 15 females;aged 31(25,43)years]and 36 patients with AQP4 antibody positive NMOSD[3 males and 33 females;aged 42(29,54)years]were collected retrospectively from September 2018 to June 2021 in Chenzhou First People′s Hospital and the Affiliated Hospital of Qingdao University which were classified as MOG group and AQP4 positive group respectively.All patients underwent routine cranial MRI scanning before treatment and the location,shape and quantity of intracranial lesions were recorded.Wilcoxon rank sum test was used to compare the number of different types of lesions between the two groups.Logistic regression analysis was used to evaluate the significance of different lesions for the two diseases.Results There were 7 types of lesions with significant differences in different parts and shapes.Stepwise Logistic regression showed that cortical and juxtacortical lesions(OR=21.91,95%CI 3.09-61.69,P<0.05)and infratentorial peripheral white matter lesions(OR=10.48,95%CI 2.00-18.89,P<0.05)were the most important risk factors in the MOG group.The incidence of cortical and juxtacortical lesions in the MOG group was 51.4%(18/35),which was higher than that in the AQP4 positive group(2.8%,1/36;χ²=19.02,P<0.01).The incidence of infratentorial peripheral white matter lesions in the MOG group was 31.4%(11/35),which was higher than that in the AQP4 positive group(5.6%,2/36;χ²=6.31,P<0.05).Receiver operating characteristic(ROC)curve showed that peripheral lesions[including 6 types of lesions such as supratentorial soft meningitis,cortical encephalitis,cortical and juxtacortical lesions,infratentorial soft meningitis,infratentorial soft meningeal demyelination and infratentorial peripheral lesions,area under curve(AUC)=0.93]were more important than cortical and juxtacortical lesions(AUC=0.75)and central lesions(supratentorial paraventricular white matter lesions,diencephalon,infratentorial paraventricular lesions,AUC=0.64),which had higher diagnostic efficiency.Conclusions The incidence of intracranial lesions in MOG antibody related demyelinating disease was higher than that in AQP4 positive NMOSD,and the distribution and morphology of intracranial MRI lesions in the two diseases had their characteristic manifestations.Identifying the distribution patterns of peripheral lesions(distributed along pia mater)and central lesions(distributed along ependyma)had a certain reference significance for distinguishing the two groups of diseases.