TY - JOUR
T1 - Common and differential connectivity profiles of deep brain stimulation and capsulotomy in refractory obsessive-compulsive disorder
AU - Chen, Xiaoyu
AU - Wang, Zhen
AU - Lv, Qian
AU - Lv, Qiming
AU - van Wingen, Guido
AU - Fridgeirsson, Egill Axfjord
AU - Denys, Damiaan
AU - Voon, Valerie
AU - Wang, Zheng
N1 - Funding Information: We specially thank Jinqiang Peng for his support to the management of our database and computation resources, where all the imaging and clinical data were harmonized and analyzed. This work was supported by the Key-Area Research and Development Program of Guangdong Province (2019B030335001), National Key R&D Program of China (No. 2017YFC1310400; No. 2018YFC1313803), Strategic Priority Research Program of Chinese Academy of Science (No. XDB32000000), grants from National Natural Science Foundation (81527901, 31771174), and Shanghai Municipal Science and Technology Major Project (No. 2018SHZDZX05). Publisher Copyright: © 2021, The Author(s), under exclusive licence to Springer Nature Limited.
PY - 2021
Y1 - 2021
N2 - Neurosurgical interventions including deep brain stimulation (DBS) and capsulotomy have been demonstrated effective for refractory obsessive-compulsive disorder (OCD), although treatment-shared/-specific network mechanisms remain largely unclear. We retrospectively analyzed resting-state fMRI data from three cohorts: a cross-sectional dataset of 186 subjects (104 OCD and 82 healthy controls), and two longitudinal datasets of refractory patients receiving ventral capsule/ventral striatum DBS (14 OCD) and anterior capsulotomy (27 OCD). We developed a machine learning model predictive of OCD symptoms (indexed by the Yale–Brown Obsessive Compulsive Scale, Y-BOCS) based on functional connectivity profiles and used graphic measures of network communication to characterize treatment-induced profile changes. We applied a linear model on 2 levels treatments (DBS or capsulotomy) and outcome to identify whether pre-surgical network communication was associated with differential treatment outcomes. We identified 54 functional connectivities within fronto-subcortical networks significantly predictive of Y-BOCS score in patients across 3 independent cohorts, and observed a coexisting pattern of downregulated cortico-subcortical and upregulated cortico-cortical network communication commonly shared by DBS and capsulotomy. Furthermore, increased cortico-cortical communication at ventrolateral and centrolateral prefrontal cortices induced by DBS and capsulotomy contributed to improvement of mood and anxiety symptoms, respectively (p < 0.05). Importantly, pretreatment communication of ventrolateral and centrolateral prefrontal cortices were differentially predictive of mood and anxiety improvements by DBS and capsulotomy (effect sizes = 0.45 and 0.41, respectively). These findings unravel treatment-shared and treatment-specific network characteristics induced by DBS and capsulotomy, which may facilitate the search of potential evidence-based markers for optimally selecting among treatment options for a patient.
AB - Neurosurgical interventions including deep brain stimulation (DBS) and capsulotomy have been demonstrated effective for refractory obsessive-compulsive disorder (OCD), although treatment-shared/-specific network mechanisms remain largely unclear. We retrospectively analyzed resting-state fMRI data from three cohorts: a cross-sectional dataset of 186 subjects (104 OCD and 82 healthy controls), and two longitudinal datasets of refractory patients receiving ventral capsule/ventral striatum DBS (14 OCD) and anterior capsulotomy (27 OCD). We developed a machine learning model predictive of OCD symptoms (indexed by the Yale–Brown Obsessive Compulsive Scale, Y-BOCS) based on functional connectivity profiles and used graphic measures of network communication to characterize treatment-induced profile changes. We applied a linear model on 2 levels treatments (DBS or capsulotomy) and outcome to identify whether pre-surgical network communication was associated with differential treatment outcomes. We identified 54 functional connectivities within fronto-subcortical networks significantly predictive of Y-BOCS score in patients across 3 independent cohorts, and observed a coexisting pattern of downregulated cortico-subcortical and upregulated cortico-cortical network communication commonly shared by DBS and capsulotomy. Furthermore, increased cortico-cortical communication at ventrolateral and centrolateral prefrontal cortices induced by DBS and capsulotomy contributed to improvement of mood and anxiety symptoms, respectively (p < 0.05). Importantly, pretreatment communication of ventrolateral and centrolateral prefrontal cortices were differentially predictive of mood and anxiety improvements by DBS and capsulotomy (effect sizes = 0.45 and 0.41, respectively). These findings unravel treatment-shared and treatment-specific network characteristics induced by DBS and capsulotomy, which may facilitate the search of potential evidence-based markers for optimally selecting among treatment options for a patient.
UR - http://www.scopus.com/inward/record.url?scp=85118192358&partnerID=8YFLogxK
U2 - https://doi.org/10.1038/s41380-021-01358-w
DO - https://doi.org/10.1038/s41380-021-01358-w
M3 - Article
C2 - 34703025
SN - 1359-4184
JO - Molecular psychiatry
JF - Molecular psychiatry
ER -