fMRI network correlates of predisposing risk factors for delirium: A cross-sectional study

Abstract

Delirium, the clinical expression of acute encephalopathy, is a common neuropsychiatric syndrome that is related to poor outcomes, such as long-term cognitive impairment. Disturbances of functional brain networks are hypothesized to predispose for delirium. The aim of this study in non-delirious elderly individuals was to investigate whether predisposing risk factors for delirium are associated with fMRI network characteristics that have been observed during delirium. As predisposing risk factors, we studied age, alcohol misuse, cognitive impairment, depression, functional impairment, history of transient ischemic attack or stroke, and physical status. In this multicenter study, we included 554 subjects and analyzed resting-state fMRI data from 222 elderly subjects (63% male, age range: 65-85 year) after rigorous motion correction. Functional network characteristics were analyzed and based on the minimum spanning tree (MST). Global functional connectivity strength, network efficiency (MST diameter) and network integration (MST leaf fraction) were analyzed, as these measures were altered during delirium in previous studies. Linear regression analyses were used to investigate the relation between predisposing delirium risk factors and delirium-related fMRI characteristics, adjusted for confounding and multiple testing. Predisposing risk factors for delirium were not associated with delirium-related fMRI network characteristics. Older age within our elderly cohort was related to global functional connectivity strength (β = 0.182, p < 0.05), but in the opposite direction than hypothesized. Delirium-related functional network impairments can therefore not be considered as the common mechanism for predisposition for delirium.

Keywords: Aging; Brain networks; Delirium; Functional connectivity; Graph theory; Risk factors.

Fig. 1

Schematic representation of the minimum spanning tree. Minimum spanning trees can conceptually range between a path-like tree (a less efficient and sparsely integrated network) and a star-like tree (an highly efficient and highly integrated network). Diameter is the length of the path between the two nodes that are furthest apart, and a measure for network efficiency. Leaf fraction is the fraction of leaf nodes (red), i.e. nodes that only have one edge, and therefore a measure of network integration. (A) Line-like network: few leaf nodes and a long diameter, (B) hierarchical tree structure: conceptually optimal topology, (C) star-like network: many leaf nodes + short diameter, central node (purple) will easily be overloaded with information. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

The relationship between age and global functional connectivity strength. A significant association was found between age and global functional connectivity strength. The association was independent of other risk factors for delirium, center, intelligence quotient or gender.