Pathological Networking of Gray Matter Dendritic Density With Classic Brain Morphometries in OCD
Xiaochen Zhang,Jiajia Zhou,Yongjun Chen,Lei Guo,Zhi Yang,Trevor W Robbins,Qing Fan
DOI: https://doi.org/10.1001/jamanetworkopen.2023.43208
2023-11-01
Abstract:Importance: The pathogenesis of obsessive-compulsive disorder (OCD) may involve altered dendritic morphology, but in vivo imaging of neurite morphology in OCD remains limited. Such changes must be interpreted functionally within the context of the multimodal neuroimaging approach to OCD. Objective: To examine whether dendritic morphology is altered in patients with OCD compared with healthy controls (HCs) and whether such alterations are associated with other brain structural metrics in pathological networks. Design, setting, and participants: This case-control study used cross-sectional data, including multimodal brain images and clinical symptom assessments, from 108 patients with OCD and 108 HCs from 2014 to 2017. Patients with OCD were recruited from Shanghai Mental Health Center, Shanghai, China, and HCs were recruited via advertisements. The OCD group comprised unmedicated adults with a Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) (DSM-IV) diagnosis of OCD, while the HCs were adults without any DSM-IV diagnosis, matched for age, sex, and education level. Data were analyzed from September 2019 to April 2023. Exposure: DSM-IV diagnosis of OCD. Main outcomes and measures: Multimodal brain imaging was used to compare neurite microstructure and classic morphometries between patients with OCD and HCs. The whole brain was searched to identify regions exhibiting altered morphology in patients with OCD and explore the interplay between the brain metrics representing these alterations. Brain-symptom correlations were analyzed, and the performance of different brain metric configurations were evaluated in distinguishing patients with OCD from HCs. Results: Among 108 HCs (median [IQR] age, 26 [23-31] years; 50 [46%] female) and 108 patients with OCD (median [IQR] age, 26 [24-31] years; 46 [43%] female), patients with OCD exhibited deficient neurite density in the right lateral occipitoparietal regions (peak t = 3.821; P ≤ .04). Classic morphometries also revealed widely-distributed alterations in the brain (peak t = 4.852; maximum P = .04), including the prefrontal, medial parietal, cingulate, and fusiform cortices. These brain metrics were interconnected into a pathological brain network associated with OCD symptoms (global strength: HCs, 0.253; patients with OCD, 0.941; P = .046; structural difference, 0.572; P < .001). Additionally, the neurite density index exhibited high discriminatory power in distinguishing patients with OCD from HCs (accuracy, ≤76.85%), and the entire pathological brain network also exhibited excellent discriminative classification properties (accuracy, ≤82.87%). Conclusions and relevance: The findings of this case-control study underscore the utility of in vivo imaging of gray matter dendritic density in future OCD research and the development of neuroimaging-based biomarkers. They also endorse the concept of connectopathy, providing a potential framework for interpreting the associations among various OCD symptom-related morphological anomalies.