The study aims to investigate whether there is difference in pre-treatment white matter parameters in treatment-resistant and treatment-responsive schizophrenia. Diffusion tensor imaging (DTI) was acquired from 60 first-episode drug-naïve schizophrenia (39 treatment-responsive and 21 treatment-resistant schizophrenia patients) and 69 age- and gender-matched healthy controls. Imaging data was preprocessed via FSL software, then diffusion parameters including fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) were extracted. Besides, structural network matrix was constructed based on deterministic fiber tracking. The differences of diffusion parameters and topology attributes between three groups were analyzed using analysis of variance (ANOVA). Compared with healthy controls, treatment-responsive schizophrenia showed altered white matter mainly in anterior thalamus radiation, splenium of corpus callosum, cingulum bundle as well as superior longitudinal fasciculus. While treatment-resistant schizophrenia patients showed white matter abnormalities in anterior thalamus radiation, cingulum bundle, fornix and pontine crossing tract relative to healthy controls. Treatment-resistant schizophrenia showed more severe white matter abnormalities in anterior thalamus radiation compared with treatment-responsive patients. There was no significant difference in white matter network topological attributes among the three groups. The performance of support vector machine (SVM) showed accuracy of 63.37% in separating the two patient subgroups ( P = 0.04). In this study, we showed different patterns of white matter alterations in treatment-responsive and treatment-resistant schizophrenia compared with healthy controls before treatment, which may help guiding patient identification, targeted treatment and prognosis improvement at baseline drug-naïve state.
本文旨在探索首发未用药状态下，抗精神病药物治疗不同反应的精神分裂症患者白质参数是否存在不同模式的改变。采集 60 例首发未用药精神分裂症患者（药物治疗有效者 39 例，难治者 21 例）与 69 例年龄、性别匹配的健康对照的磁共振弥散张量成像（DTI）数据，运用 FSL 软件进行数据处理，提取反映白质微结构完整性的参数并进行确定性的纤维追踪，并基于纤维追踪的结果构建网络矩阵，通过方差分析比较三组间白质参数及白质连接网络拓扑属性的差异。结果发现相比于健康对照，药物治疗有效组白质异常主要表现在丘脑前辐射、胼胝体压部、扣带束和上纵束，而药物难治组白质异常主要分布在丘脑前辐射、扣带束、穹隆和脑桥交叉纤维束。药物难治组在丘脑前辐射表现出比药物治疗有效组更严重的白质异常。此外，三组被试之间白质连接网络拓扑属性差异无统计学意义。支持向量机（SVM）的结果显示两个患者亚组之间分类的准确度为 63.37%（ P = 0.04）。该研究表明精神分裂症抗精神病药物治疗不同反应者存在不同模式的白质微结构完整性损伤，这对基线未用药状态下区分和治疗不同反应患者、改善患者预后具有指导作用。.
Keywords: diffusion tensor imaging; schizophrenia; tract-based spatial statistics; treatment outcome.