ReviewResting-state functional connectivity in major depressive disorder: A review
Introduction
Major depressive disorder (MDD) is one of the most prevalent psychiatric disorders and is the second leading cause of disability worldwide (Ferrari et al., 2013). Despite almost 60 years of intensive neurobiological research, our current understanding of its pathophysiology is limited, which is reflected in a heterogeneous disease concept and moderate effects of treatment (Arroll et al., 2009, Mojtabai, 2013, Moncrieff et al., 2004). While earlier neuroimaging techniques have investigated focal structural and functional changes (Alcaro et al., 2010, Hamilton et al., 2013, Koenigs and Grafman, 2009, Northoff et al., 2011, Price and Drevets, 2012), depression is increasingly understood as a disorder of distributed effects of aberrant interaction in the brain (Drevets et al., 2008, Hamilton et al., 2013, Mayberg, 1997). Within this framework, brain regions are dynamically organized into functional networks of interconnected areas (or “nodes”) that interact to perform specific tasks (Bressler, 1995).
With the advance of network-based research in system-level neurosciences in general, new techniques allow us to identify these large-scale brain networks, for example by looking at changes in blood-oxygen-level-dependent (BOLD) signal using functional magnetic resonance imaging (fMRI). An important methodological development in investigating these networks was the finding that they can consistently be identified during the “resting-state”, i.e. when a subject is not engaged in any particular task (Biswal et al., 1995). This independence of task-based paradigms offers the important advantage of being reproducible across different populations and study settings. Following this, many recent studies have investigated how the different nodes and networks communicate by investigating synchronous spontaneous activity in different regions of the brain. This so-called “resting-state functional connectivity” (hereafter referred to as “connectivity”) represents the temporal coherence of the BOLD-signal within or between regions or networks during rest and is an important addition to functional imaging techniques in unraveling the neurobiology of depression (Friston, 2011).
In MDD most findings in task-based and resting-state fMRI implicate one of three major neural networks: the default mode network (DMN), the central executive network (CEN) and the salience network (SN) (Hamilton et al., 2013, Menon, 2011, Raichle et al., 2001, Seeley et al., 2007). Two recent papers have reviewed studies on functional connectivity in depression (Smith, 2014, Wang et al., 2012), but either included only a limited number of eligible studies (Smith, 2014) or rather divergent methods which makes comparison of results problematic (Wang et al., 2012). Because of these limitations, and the large number of connectivity papers published in recent years, we aim to provide a coherent review of the resting-state functional connectivity literature in depression that takes into account the different methods used, and update the current concept of depression as a network-based disorder.
Our review will focus on changes within (1) the default mode network, (2) the central executive network, (3) the salience network and (4) the interactions between these networks. We will start by giving a short overview of these networks and their function. Then, we will provide a critical appraisal of all available studies on resting-state functional connectivity in MDD, taking into account different methods used as well as the relation to clinical characteristics and effects of treatment. Lastly, we will discuss the significance of these findings in the light of current depression hypotheses.
Section snippets
Core large-scale networks in major depressive disorder
The default mode network (DMN), the central executive network (CEN) and the salience network (SN) (Fig. 1) (Hamilton et al., 2013, Menon, 2011, Raichle et al., 2001, Seeley et al., 2007) represent the brain's function during rest, cognition and emotional processes, all of which are essential processes that are altered in depression.
Literature search
A search in PubMed/MedLine was performed to identify papers in English reporting on resting-state functional connectivity in unipolar MDD, using “major depressive disorder”, “resting state” and “functional connectivity” as search terms. Papers were selected when they (1) included patients with current MDD, (2) used fMRI, (3) made comparisons to matched healthy controls and (4) reported on measures of functional connectivity. References of the included papers were checked for citations that were
Results
In total 8 papers using ICA (Table 1) and 28 papers applying SCA (Table 2) were included. Papers were published between 2005 and 2014, but 34 out of 36 studies were reported in the past five years. All ICA papers have investigated the DMN, while the CEN and SN were only reported in a subset of studies. Among SCA papers, the most frequently selected seed regions are the ACC, PCC and the amygdala. 25 out of the 36 studies investigated an adult population, 6 looked at an elderly group and 5
Discussion
Due to the relative ease of collecting resting-state fMRI scans and the availability of new techniques such as ICA to accommodate the large amount of information involved, a large body of investigations into resting-state functional connectivity in depression is available now. Our review of the currently available data has yielded several findings that were consistent across different methods:
- (1)
increased connectivity within the anterior DMN,
- (2)
increased connectivity between the anterior DMN and the
Conclusion
Connectivity studies in MDD expand upon activation studies by reporting increased connectivity within nodes of the anterior DMN and between the anterior DMN and the SN, changed connectivity between the anterior and posterior DMN and decreased connectivity between the posterior DMN and the CEN. We propose that this reflects a state of increased interaction between self-referential and emotional networks, and the dominance of negative self-referential over cognitive processing which corresponds
References (109)
- et al.
Is subcortical–cortical midline activity in depression mediated by glutamate and GABA?. A cross-species translational approach
Neurosci. Biobehav. Rev.
(2010) - et al.
Functional connectivity in the cognitive control network and the default mode network in late-life depression
J. Affect. Disord.
(2012) - et al.
Activity and connectivity of brain mood regulating circuit in depression: a functional magnetic resonance study
Biol. Psychiatry
(2005) - et al.
Resting state functional connectivity and treatment response in late-life depression
Psychiatry Res.
(2013) - et al.
Functional-anatomic fractionation of the brain's default network
Neuron
(2010) Modelling with independent components
NeuroImage
(2012)Large-scale cortical networks and cognition
Brain Res. Rev.
(1995)- et al.
The predictive power of subgroups: an empirical approach to identify depressive symptom patterns that predict response to treatment
J. Affect. Disord.
(2014) - et al.
Disrupted resting-state functional connectivity of the hippocampus in medication-naïve patients with major depressive disorder
J. Affect. Disord.
(2012) - et al.
Resting-state functional connectivity of subgenual anterior cingulate cortex in depressed adolescents
Biol. Psychiatry
(2013)