TY - JOUR

T1 - Patterns of Cortical Structures and Cognition in Antipsychotic-Naïve Patients With First-Episode Schizophrenia

T2 - A Partial Least Squares Correlation Analysis

AU - Jessen, Kasper

AU - Mandl, Rene C.W.

AU - Fagerlund, Birgitte

AU - Bojesen, Kirsten B.

AU - Raghava, Jayachandra M.

AU - Obaid, Hayder G.

AU - Jensen, Marie B.

AU - Johansen, Louise B.

AU - Nielsen, Mette

AU - Pantelis, Christos

AU - Rostrup, Egill

AU - Glenthøj, Birte Y.

AU - Ebdrup, Bjørn H.

PY - 2019

Y1 - 2019

N2 - Background: Schizophrenia is associated with alterations in cortical structures and cognitive impairments, but antipsychotic medication may affect these measures. We investigated patterns of relationships between cortical structures and cognitive domains in antipsychotic-naïve patients with first-episode schizophrenia. Methods: T1-weighted 3T magnetic resonance imaging was performed in 105 patients and 136 healthy control subjects. Using FreeSurfer, we obtained measurements of cortical thickness, surface area, and mean curvature. Using an extensive neurocognitive battery including the Danish Adult Reading Test and subtests from the Cambridge Neuropsychological Test Automated Battery, we obtained estimates of premorbid intelligence, spatial working memory, spatial planning, intra-extradimensional set shifting, and reaction and movement times. With univariate analyses, we tested group differences between cortical structures and cognition. With partial least squares correlation analyses, we investigated patterns of associations between cortical structures and cognition. Results: Patients had significantly higher mean curvature and were impaired on 7 of 11 cognitive parameters. The between-group partial least squares correlation analysis revealed two cortical thickness/cognition patterns that differentiated patients and healthy control subjects (omnibus test, p =.011). Most cortical regions contributed reliably to these patterns. In patients, spatial working memory, spatial planning, reaction and movement times, and premorbid intelligence contributed reliably to the pattern; in healthy control subjects, spatial planning and intra-extradimensional set shifting contributed reliably. Conclusions: Antipsychotic-naïve patients with first-episode schizophrenia displayed a higher mean curvature, but no significant difference in other gray matter indices was found. Nevertheless, the pattern of associations between global cortical thickness and cognitive functions was markedly different between groups. These multivariate analyses reveal a novel linkage between regional cortical brain structure and cognitive deficits at the earliest, never-medicated illness stage.

AB - Background: Schizophrenia is associated with alterations in cortical structures and cognitive impairments, but antipsychotic medication may affect these measures. We investigated patterns of relationships between cortical structures and cognitive domains in antipsychotic-naïve patients with first-episode schizophrenia. Methods: T1-weighted 3T magnetic resonance imaging was performed in 105 patients and 136 healthy control subjects. Using FreeSurfer, we obtained measurements of cortical thickness, surface area, and mean curvature. Using an extensive neurocognitive battery including the Danish Adult Reading Test and subtests from the Cambridge Neuropsychological Test Automated Battery, we obtained estimates of premorbid intelligence, spatial working memory, spatial planning, intra-extradimensional set shifting, and reaction and movement times. With univariate analyses, we tested group differences between cortical structures and cognition. With partial least squares correlation analyses, we investigated patterns of associations between cortical structures and cognition. Results: Patients had significantly higher mean curvature and were impaired on 7 of 11 cognitive parameters. The between-group partial least squares correlation analysis revealed two cortical thickness/cognition patterns that differentiated patients and healthy control subjects (omnibus test, p =.011). Most cortical regions contributed reliably to these patterns. In patients, spatial working memory, spatial planning, reaction and movement times, and premorbid intelligence contributed reliably to the pattern; in healthy control subjects, spatial planning and intra-extradimensional set shifting contributed reliably. Conclusions: Antipsychotic-naïve patients with first-episode schizophrenia displayed a higher mean curvature, but no significant difference in other gray matter indices was found. Nevertheless, the pattern of associations between global cortical thickness and cognitive functions was markedly different between groups. These multivariate analyses reveal a novel linkage between regional cortical brain structure and cognitive deficits at the earliest, never-medicated illness stage.

KW - Antipsychotic-naïve

KW - Cognition

KW - Cortical brain structure

KW - First-episode schizophrenia

KW - Partial least squares correlation

U2 - 10.1016/j.bpsc.2018.09.006

DO - 10.1016/j.bpsc.2018.09.006

M3 - Journal article

C2 - 30420252

AN - SCOPUS:85056221180

VL - 4

SP - 444

EP - 453

JO - Biological Psychiatry: Cognitive Neuroscience and Neuroimaging

JF - Biological Psychiatry: Cognitive Neuroscience and Neuroimaging

SN - 2451-9022

IS - 5

ER -