Transient Patterns of Functional Dysconnectivity in Clinical High Risk and Early Illness Schizophrenia Individuals Compared with Healthy Controls

Eva Mennigen, Susanna L. Fryer, Barnaly Rashid, Eswar Damaraju, Yuhui Du, Rachel L. Loewy, Barbara K. Stuart, Vince Daniel Calhoun, Daniel H. Mathalon

Research output: Contribution to journalArticle

Abstract

Schizophrenia shows abnormal dynamic functional network connectivity (dFNC), but it is unclear whether these abnormalities are present early in the illness course or precede illness onset in individuals at clinical high risk (CHR) for psychosis. We examined dFNC from resting-state functional magnetic resonance imaging data in CHR (n = 53), early illness schizophrenia (ESZ; n = 58), and healthy control (HC; n = 70) individuals. We applied a sliding temporal window approach capturing five distinct dFNC states. In ESZ patients, the likelihood of transitioning from state 4, a state that exhibited greater cortical-subcortical hyperconnectivity and also lacked typically observed anticorrelation between the default mode network and other functional networks, to a hypoconnected state was increased compared with HC and CHR groups. Furthermore, we investigated the interaction of group and state on dFNC. Overall, HC individuals showed significant changes of connectivity between states that were absent or altered in ESZ patients and CHR individuals. Connectivity differences between groups were identified primarily in two out of the five states, in particular, between HC and ESZ groups. In summary, it appears that the interaction effect was mostly driven by (1) dynamic connectivity changes in HC that were abnormal in CHR and ESZ individuals and (2) the fact that dysconnectivity between groups was only present in some states. These findings underscore the likelihood that abnormalities are present not only in static FNC but also in dFNC, in individuals at CHR for schizophrenia.

Original languageEnglish (US)
Pages (from-to)60-76
Number of pages17
JournalBrain connectivity
Volume9
Issue number1
DOIs
StatePublished - Feb 1 2019
Externally publishedYes

Fingerprint

Schizophrenia
Psychotic Disorders
Magnetic Resonance Imaging

Keywords

  • clinical high risk
  • dysconnectivity
  • independent component analysis
  • schizophrenia

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Transient Patterns of Functional Dysconnectivity in Clinical High Risk and Early Illness Schizophrenia Individuals Compared with Healthy Controls. / Mennigen, Eva; Fryer, Susanna L.; Rashid, Barnaly; Damaraju, Eswar; Du, Yuhui; Loewy, Rachel L.; Stuart, Barbara K.; Calhoun, Vince Daniel; Mathalon, Daniel H.

In: Brain connectivity, Vol. 9, No. 1, 01.02.2019, p. 60-76.

Research output: Contribution to journalArticle

Mennigen, Eva ; Fryer, Susanna L. ; Rashid, Barnaly ; Damaraju, Eswar ; Du, Yuhui ; Loewy, Rachel L. ; Stuart, Barbara K. ; Calhoun, Vince Daniel ; Mathalon, Daniel H. / Transient Patterns of Functional Dysconnectivity in Clinical High Risk and Early Illness Schizophrenia Individuals Compared with Healthy Controls. In: Brain connectivity. 2019 ; Vol. 9, No. 1. pp. 60-76.
@article{69c8f00a1be048bd9cd9a14fea393dc1,
title = "Transient Patterns of Functional Dysconnectivity in Clinical High Risk and Early Illness Schizophrenia Individuals Compared with Healthy Controls",
abstract = "Schizophrenia shows abnormal dynamic functional network connectivity (dFNC), but it is unclear whether these abnormalities are present early in the illness course or precede illness onset in individuals at clinical high risk (CHR) for psychosis. We examined dFNC from resting-state functional magnetic resonance imaging data in CHR (n = 53), early illness schizophrenia (ESZ; n = 58), and healthy control (HC; n = 70) individuals. We applied a sliding temporal window approach capturing five distinct dFNC states. In ESZ patients, the likelihood of transitioning from state 4, a state that exhibited greater cortical-subcortical hyperconnectivity and also lacked typically observed anticorrelation between the default mode network and other functional networks, to a hypoconnected state was increased compared with HC and CHR groups. Furthermore, we investigated the interaction of group and state on dFNC. Overall, HC individuals showed significant changes of connectivity between states that were absent or altered in ESZ patients and CHR individuals. Connectivity differences between groups were identified primarily in two out of the five states, in particular, between HC and ESZ groups. In summary, it appears that the interaction effect was mostly driven by (1) dynamic connectivity changes in HC that were abnormal in CHR and ESZ individuals and (2) the fact that dysconnectivity between groups was only present in some states. These findings underscore the likelihood that abnormalities are present not only in static FNC but also in dFNC, in individuals at CHR for schizophrenia.",
keywords = "clinical high risk, dysconnectivity, independent component analysis, schizophrenia",
author = "Eva Mennigen and Fryer, {Susanna L.} and Barnaly Rashid and Eswar Damaraju and Yuhui Du and Loewy, {Rachel L.} and Stuart, {Barbara K.} and Calhoun, {Vince Daniel} and Mathalon, {Daniel H.}",
year = "2019",
month = "2",
day = "1",
doi = "10.1089/brain.2018.0579",
language = "English (US)",
volume = "9",
pages = "60--76",
journal = "Brain Connectivity",
issn = "2158-0014",
publisher = "Mary Ann Liebert Inc.",
number = "1",

}

TY - JOUR

T1 - Transient Patterns of Functional Dysconnectivity in Clinical High Risk and Early Illness Schizophrenia Individuals Compared with Healthy Controls

AU - Mennigen, Eva

AU - Fryer, Susanna L.

AU - Rashid, Barnaly

AU - Damaraju, Eswar

AU - Du, Yuhui

AU - Loewy, Rachel L.

AU - Stuart, Barbara K.

AU - Calhoun, Vince Daniel

AU - Mathalon, Daniel H.

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Schizophrenia shows abnormal dynamic functional network connectivity (dFNC), but it is unclear whether these abnormalities are present early in the illness course or precede illness onset in individuals at clinical high risk (CHR) for psychosis. We examined dFNC from resting-state functional magnetic resonance imaging data in CHR (n = 53), early illness schizophrenia (ESZ; n = 58), and healthy control (HC; n = 70) individuals. We applied a sliding temporal window approach capturing five distinct dFNC states. In ESZ patients, the likelihood of transitioning from state 4, a state that exhibited greater cortical-subcortical hyperconnectivity and also lacked typically observed anticorrelation between the default mode network and other functional networks, to a hypoconnected state was increased compared with HC and CHR groups. Furthermore, we investigated the interaction of group and state on dFNC. Overall, HC individuals showed significant changes of connectivity between states that were absent or altered in ESZ patients and CHR individuals. Connectivity differences between groups were identified primarily in two out of the five states, in particular, between HC and ESZ groups. In summary, it appears that the interaction effect was mostly driven by (1) dynamic connectivity changes in HC that were abnormal in CHR and ESZ individuals and (2) the fact that dysconnectivity between groups was only present in some states. These findings underscore the likelihood that abnormalities are present not only in static FNC but also in dFNC, in individuals at CHR for schizophrenia.

AB - Schizophrenia shows abnormal dynamic functional network connectivity (dFNC), but it is unclear whether these abnormalities are present early in the illness course or precede illness onset in individuals at clinical high risk (CHR) for psychosis. We examined dFNC from resting-state functional magnetic resonance imaging data in CHR (n = 53), early illness schizophrenia (ESZ; n = 58), and healthy control (HC; n = 70) individuals. We applied a sliding temporal window approach capturing five distinct dFNC states. In ESZ patients, the likelihood of transitioning from state 4, a state that exhibited greater cortical-subcortical hyperconnectivity and also lacked typically observed anticorrelation between the default mode network and other functional networks, to a hypoconnected state was increased compared with HC and CHR groups. Furthermore, we investigated the interaction of group and state on dFNC. Overall, HC individuals showed significant changes of connectivity between states that were absent or altered in ESZ patients and CHR individuals. Connectivity differences between groups were identified primarily in two out of the five states, in particular, between HC and ESZ groups. In summary, it appears that the interaction effect was mostly driven by (1) dynamic connectivity changes in HC that were abnormal in CHR and ESZ individuals and (2) the fact that dysconnectivity between groups was only present in some states. These findings underscore the likelihood that abnormalities are present not only in static FNC but also in dFNC, in individuals at CHR for schizophrenia.

KW - clinical high risk

KW - dysconnectivity

KW - independent component analysis

KW - schizophrenia

UR - http://www.scopus.com/inward/record.url?scp=85061982607&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061982607&partnerID=8YFLogxK

U2 - 10.1089/brain.2018.0579

DO - 10.1089/brain.2018.0579

M3 - Article

C2 - 29855202

AN - SCOPUS:85061982607

VL - 9

SP - 60

EP - 76

JO - Brain Connectivity

JF - Brain Connectivity

SN - 2158-0014

IS - 1

ER -