F82. INDIVIDUAL GRAY MATTER NETWORKS AND INSIGHT IN PSYCHOTIC DISORDERS. (9th April 2019)

• Record Type: Journal Article
• Authors: Larabi, Daouia; Marsman, Jan-Bernard; Aleman, Andre; Tijms, Betty; Opmeer, Esther; Pijnenborg, Marieke (Gerdina); van der Meer, Lisette; Van Tol, Marie-Jose; Curcic-Blake, Branislava
• Is Part Of: Schizophrenia bulletin. Volume 45(2019)Supplement 2
• Page Start: S285
• Page End: S285
• Publication Date: 2019-04-09
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbz018.494
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: Clinical insight is impaired in the majority of individuals with schizophrenia (Dam, 2006) and is associated with poorer outcome (Lincoln et al., 2007). Impaired insight cannot be pinpointed to abnormalities of isolated brain areas, as earlier studies on cortical structure and insight found abnormalities in a distributed network of brain regions (Pijnenborg et al., in prep). Getting a better understanding of the neural substrate of impaired insight might help in finding better treatment options to improve insight. Typically, brains are characterized by small-world topology, reflected by a balance between information segregation (i.e. short distances between nodes) and integration (i.e. high clustering of nodes). Several studies have shown less characteristics of small-world topology in patients with schizophrenia. In this study, we used tools of graph theory to investigate whether there are less characteristics of small-world topology of individual structural networks in patients with a psychotic disorder compared to healthy individuals, and whether this is related to interindividual differences in insight. Methods: T1-weighted images of 114 patients with a psychotic disorder (76% males; mean age=33.67, SD=10.86) and 54 healthy controls (63% males; mean age=35.11, SD=10.76) were acquired with a 3T Philips Intera MRI-scanner. Clinical insight was measured with item G12 of the PANSS (Kay et al., 1987). In a subsample of 62 patients, clinical insight was also measured with the Schedule of Assessment of Insight – Expanded (SAI-E; Kemp & David, 1997), and cognitive insight with the Beck Cognitive Insight Scale (BCIS; Beck et al., 2004). Brain images were segmented using SPM12. Individual gray matter similarity networks were created from gray matter segmentations using a previously described method (Tijms et al., 2012). We calculated the following graph metrics: path length, clustering coefficient, betweenness centrality, lambda (i.e. normalized path length), gamma (i.e. normalized clustering coefficient) and small-world property (i.e. gamma/lambda) using the Brain Connectivity Toolbox.8 Differences in graph metrics between patients and healthy individuals were investigated with ANCOVA's. The associations between insight and graph metrics were calculated with partial correlations. Education and total gray matter were entered as covariates in all analyses. Results: We found significantly lower clustering coefficient (F(1, 165)=17.90, pFDR<0.001) and higher betweenness centrality (F(1, 165)=7.910, punc=0.006, pFDR=0.018) in patients compared to healthy controls. In addition, we found that poorer ability to relabel symptoms (SAI-E subscale) correlated with higher betweenness centrality (rs=-0.359, punc=0.005, pbonf=0.04). Discussion: Our result of less clustering indicates less effective integration of information and a more random topology of networks of patients with a psychotic disorder. Our findings of higher betweenness centrality values in patients compared to healthy individuals, and higher betweenness centrality values in patients with poorer ability to attribute symptoms to the illness indicate increased hub-characteristics of regions in these patients. Hubs are central regions that interact with many other regions and facilitate integration. Hub regions make graphs more resilient toward pathological damage but are also the weakest points of these networks. We will further investigate this relationship by examining which brain regions show this relationship specifically. Learning more about abnormalities of underlying structural networks by examining small-world property may help in getting a better understanding of impaired insight and finding potential biomarkers.