摘要目的 探讨注意缺陷多动障碍(ADHD)患儿脑结构网络连接改变,为临床早期识别ADHD提供新的标志物.方法 对2017年5月至2018年5月徐州医科大学附属徐州儿童医院25例ADHD患儿和23例健康对照儿童通过确定性弥散张量示踪法获得脑结构网络,利用图论分析技术在全局属性、节点属性及网络连边的水平评估2组间的网络结构连接差异.结果 (1)ADHD组全局效率(0.30±0.13)明显低于健康对照组(0.38±0.11),聚类系数(0.35±0.28)和特征路径长度(2.94±0.38)明显高于健康对照组(0.28 ±0.10、2.65 ±0.37),差异均有统计学意义(t=-2.41、2.31、2.62,均P＜0.05).(2)ADHD组左侧三角部额下回(0.13±0.06)、缘上回(0.30±0.10)、顶下缘角回(0.29±0.10)及楔前叶(0.26±0.12)节点效率明显低于健康对照组(0.17 ±0.07、0.38±0.10、0.40 ±0.12、0.35 ±0.12),而右侧眶部额上回(0.61±0.19)和中央旁小叶(0.54±0.13)节点效率明显高于健康对照组(0.49 ±0.17、0.43 ±0.14),差异均有统计学意义[t=-2.52、-2.62、-3.11、-2.77、2.34、2.79,均P＜0.05,伪发现率(FDR)校正].(3)ADHD组观察到由左侧额顶区、基底核、丘脑和岛叶构成连接破坏的子网络(P＜0.05,FDR校正),且改变的子网络的连接强度能较好地将ADHD患儿与健康对照儿童区分开(受试者工作特征曲线下面积为0.78).(4)子网络连接强度降低与ADHD患儿的注意缺陷有关(r=-0.607,P=0.003).结论 使用磁共振弥散张量成像,借助图论分析技术,ADHD患儿在多个水平可观察到脑结构网络的改变;脑结构网络属性和连边改变的分布和模式有望成为早期识别ADHD的新标志物.

abstractsObjective To explore the changes in brain structure network connection in children with attention deficit hyperactivity disorder(ADHD),and to provide novel markers for early identification of ADHD in clinical practice.Methods Deterministic diffusion-tensor tractography and graph theory approaches were used to investigate the topologic organization of the brain structural connectome in 25 children with ADHD and 23 healthy control children from May 2017 to May 2018,at Children's Hospital of Xuzhou Medical University.Individual white matter networks were constructed for each participant,then the global properties,nodal properties and edge-wise distributions were compared between the two groups.Results (1) The global efficiency of the ADHD group (0.30 ± 0.13) was significantly lower than that of the healthy control group (0.38 ± 0.11),but the clustering coefficient (0.35 ± 0.28) and the characteristic path length (2.94 ±0.38) were significantly higher than those of the healthy control group (0.28 ±0.10,2.65 ± 0.37),and the differences were statistically significant (t =-2.41,2.31,2.62,all P ＜ 0.05).(2) In the A DHD group,the nodal efficiency of the left inferior frontal gyrus,triangular part (0.13 ± 0.06),left supramarginal gyrus (0.30 ±0.10),left inferior parietal,angular gyri (0.29 ±0.10),left precuneus (0.26 ±0.12)were significantly lower than the healthy control group(0.17 ±0.07,0.38 ±0.10,0.40 ±0.12,0.35 ±0.12),while the nodal efficiency of the right superior frontal gyrus,orbital part and right paracentral lobule were significantly higher than the healthy control group(0.49 ± 0.17,0.43 ± 0.14),and the differences were statistically significant [t =-2.52,-2.62,-3.11,-2.77,2.34,2.79,all P ＜ 0.05,false discovery rate (FDR) corrected].(3) A disrupted subnetwork was observed that consisted of left frontoparietal areas,basal ganglia,thalamus and insular network (P ＜ 0.05,FDR corrected),which has the potential to discriminate individuals with ADHD from healthy control children (area under receiver operating characteristic curve was 0.78).(4) Diminished strength of the subnet work connections was correlated with the attention defect in patients with ADHD (r =-0.607,P =0.003).Conclusions Using magnetic resonance diffusion tensor imaging,with the help of graph theory analysis technology,ADHD children can be observed changes in brain structure network at multiple levels.The distribution pattern of brain network structure connection changes is expected to become a new marker for identifying ADHD.