F76. ALTERED PARAHIPPOCAMPAL GYRUS ACTIVATION AND ITS CONNECTIVITY WITH RESTING-STATE NETWORK AREAS IN SCHIZOPHRENIA: A QEEG STUDY. (9th April 2019)

• Record Type: Journal Article
• Authors: Soni, Sunaina; Muthukrishnan, Suriya Prakash; Sood, Mamta; Kaur, Simran; Sharma, Ratna
• Is Part Of: Schizophrenia bulletin. Volume 45(2019)Supplement 2
• Page Start: S283
• Page End: S283
• Publication Date: 2019-04-09
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbz018.488
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: Neural oscillations at rest are responsible for coordinating synchronized and coherent activity during normal brain functioning. Schizophrenia is reported as a disconnection syndrome, therefore, neural oscillations and their synchrony are a crucial target to understand the pathophysiology of schizophrenia. Many researchers have reported consistent findings of augmented low frequency power in delta and theta bands and reduced high frequency activity during rest in patients with schizophrenia. Disturbances in neural oscillations could be assessed by QEEG as an endophenotypic and state markers for early detection and management strategies in asymptomatic first-degree relatives of patients as well. Methods: Thirty-two patients with schizophrenia (18–45 years, DSM-5 criteria), 28 first-degree relatives and 31 healthy controls participated in the study. Brain activity during eyes closed condition was recorded using 128 channel EEG. After pre-processing and independent component analysis (ICA), an equivalent current dipole was estimated for each independent component (IC). Total of 1551 independent and localizable EEG components across all the groups were used in subsequent analysis. ICs were clustered by applying K-means algorithm using dipole locations, scalp maps and power spectrum. Power spectral density was computed using 'spectopo' function. Nonparametric permutation-based statistical method was applied in conjunction with the cluster correction (max method) with p-value < 0.05 to avoid the issue of multiple comparisons. Source coherence was calculated between IC clusters using 'newcrossf' function. Results: ICs were clustered into 19 clusters and 131 components were identified as outliers. 12 clusters were excluded from further analysis because they were comprised of components from less than half the total number of participants from each group and had insignificant difference between the groups on permutation tests. Seven clusters were analyzed further for group differences. Dipole locations of the cluster centroids for the clusters identified across groups at rest were right insula, left parahippocampal gyrus, cingulate gyrus, left insula, left anterior cingulate and left posterior cingulate. Patients with schizophrenia and first-degree relatives displayed significantly higher power spectral density than healthy controls for all the frequency bands at left parahippocampal gyrus (-7, -26, 8; BA 27). Patients with schizophrenia showed significantly lower power spectral density compared to first-degree relatives and healthy controls in gamma and alpha, beta and gamma frequency bands, at left parahippocampal gyrus (-4, -28, 1; BA 27). Functional connectivity was found to be decreased in patients and increased in relatives compared to healthy controls between different resting-state network areas. Discussion: Hyperactivity and hyperconnectivity in first-degree relatives compared to healthy controls can be considered as default mode and resting-state network suppression deficit at rest signifying familial vulnerability to schizophrenia suggestive of a compensatory response. Therefore, aberrant power and functional connectivity in default mode and resting-state network areas in patients with schizophrenia and their first-degree relatives can serve as endophenotypic and state markers of schizophrenia respectively.