Network-based decoupling of local gyrification in obsessive-compulsive disorder

Abstract

Gyrification is associated with cortical maturation and closely linked to neurodevelopmental processes. Obsessive-compulsive disorder has previously been associated with neurodevelopmental risk factors. Using graph theoretical modeling we examined structural covariance patterns to assess potential disruptions in processes associated with neurodevelopment in OCD. In total 97 patients and 92 healthy controls underwent magnetic resonance imaging. Structural covariance networks based on local gyrification indices were constructed using an atlas-based parcellation scheme. Network properties were assessed using the network-based statistic as well as global and local graph theoretical measures. Correlations between gyrification and symptom severity as well as age of disease onset were examined. Network-based statistic analysis revealed one cluster with significantly decreased structural covariance in patients comprising mainly ventral brain regions (p = .041). Normalized characteristic path length was found to be impaired in patients (p = .051). On a nodal level, left middle frontal sulcus displayed a significantly decreased local clustering coefficient (p < .001). Finally, gyrification in several inferior frontal nodes significantly correlated with age of onset but not symptom severity. The decrease in a gyrification-based covariance network in OCD appears to be mostly confined to ventral areas in which gyrification starts the latest during development. This pattern may indicate that alterations taking place during development are potentially time locked to specific periods. Correlations between gyrification in inferio-frontal nodes and age of onset potentially indicate a structural trait rather than state marker for OCD. Finally, a trend in impaired global integration capabilities may point towards potentially widespread global alterations during neurodevelopment in patients.

Keywords: MRI; OCD; connectome; gyrification; network.

Figure 1

Network‐based statistic analysis results. Visualization of the significantly altered gyrification‐based structural covariance network (i.e., decreased covariance) in OCD based on network‐based statistic analysis (p = .041). Red dots indicate nodes comprising the cluster, gray lines indicate edges between nodes [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

Global graph‐measures as a function of density. (a) Between‐group differences in normalized average clustering coefficients, γ. (b) Between‐group differences in normalized average characteristic path length, λ. (c) Between‐group differences in small‐world index, σ [Color figure can be viewed at http://wileyonlinelibrary.com]