Research reportAttention and executive functions in remitted major depression patients
Introduction
It is widely accepted that during an episode, patients with major depression disorder (MDD) show cognitive deficits in several domains (Elliott, 2002). Neuropsychological deficits have been demonstrated in memory, verbal and nonverbal learning, selective and sustained attention, alertness (simple reaction time tasks), and executive functions, such as cognitive flexibility, problem-solving, planning, and monitoring (Austin et al., 1992, Veiel, 1997, Zakzanis et al., 1998, Ottowitz et al., 2002).
Empirical findings regarding cognitive functions after remission of clinical symptoms are not conclusive. While a number of recent studies claim state-dependent phenomena (Austin et al., 2001), it was also possible to demonstrate persistent cognitive deficits in the domains of memory (Marcos et al., 1994, Frasch et al., 2000), attention (Trichard et al., 1995, Paradiso et al., 1997, Tham et al., 1997, Frasch et al., 2000, Weiland-Fiedler et al., 2004), and executive functions (Grant et al., 2001). Beats et al. (1996) found no deficits in remitted depressives. Kuny et al. (1997) suggest different subgroups of MDD patients: a therapy-refractory stable form, and a subgroup of MDD patients whose cognitive deficits are responsive to therapy.
One reason for the inconsistency of results lies in the differing considerations of moderator variables. Some authors, for instance, fail to distinguish between unipolar and bipolar affective disorders (Tham et al., 1997, Frasch et al., 2000), while others investigated patients with different subtypes of depression (Paradiso et al., 1997). Furthermore, varying assessments of residual symptoms and broad inclusion criteria (e.g. comorbid disorders) may have influenced the results (Elliott, 2002). Moreover, neuropsychological deficits are influenced by severity of disease and by specific symptoms, e.g. psychotic symptoms (Nelson et al., 1998). Weiland-Fiedler et al. (2004) tested frontostriatal functions and temporal and parietal lobe functions of remitted patients with recurrent MDD. The authors defined remission as a period of at least 3 months during which the patients were not medicated, with non-depressed scores on the Montgomery-Åsberg Depression Rating Scale (MADRS) (Neumann and Schulte, 1989). They found continuing disturbances in sustained attention, and more subtle disturbance in the mnemonic and strategic aspects of working memory and psychomotor functioning (Weiland-Fiedler et al., 2004). Different courses of illness as indicated by number of episodes were not taken into account in their study design. Christensen et al. (1997) indicated that performance in different tasks of attention, concentration, and mental speed has to be determined before higher cognitive functions such as problem solving are to be tested.
In depression, fronto-subcortical networks, including the dorsolateral prefrontal cortex (DLPFC), the ventrolateral prefrontal cortex (VLPFC), the anterior cingulate cortex (ACC), the thalamus, the striatum, the basal ganglia (BG), and the hippocampus are involved (Brody et al., 2001).
MDD is associated with decreased volume in dorsolateral and dorsomedial PFC, in the subgenual region of ACC, in the basal ganglia, and the hippocampus. During an episode, decreased metabolic activity has been found in the same regions. Increased levels of activity have been shown in ventrolateral and orbital PFC and in the amygdala (Davidson et al., 2002). During the depressive episode, the ACC and DLPFC receive abnormal inputs/signals from the amygdala (Drevets, 2000). It is known from activation studies in healthy subjects that these brain areas constitute part of widely distributed networks of attention and executive functions (Cabeza and Nyberg, 2000). It is therefore assumed that neuropsychological malfunctions in MDD are based on pathological projections from the amygdala to ACC and PFC (Elliott et al., 1997, Davidson et al., 2002). Drevets (2000) and Harrison (2002) reported that hyperactivation of the amygdala persists after remission. Persistent hyperactivation can chronically influence attention and executive functions.
Strakowski et al. (2000) and Strakowski et al. (2002) assume that persistent hyper- or hypoactivation of critical subcortical regions (amygdala, nucleus caudatus) in combination with fronto-subcortical networks can disturb the modulatory input to the prefrontal cortex. The pathological input could result in enduring instable hypo- or hyperactivated states. These changes are not presumed to be state dependent (Harrison, 2002, Drevets, 2000). An investigation by Kumar et al. (1998) showed a correlation between structural alterations of the PFC and severity of course of illness. As a result, disease-associated dysexecutive symptoms may persist and increase during the course of illness.
However, neuropsychological studies involving remitted MDD patients with different courses of illness are lacking. From anatomical and neurophysiological findings it can be assumed that deficits of attention and executive functions persist in MDD patients after the depressive episode is remitted. Moreover, DLPFC-associated executive impairments should be correlated to duration of disease.
Our investigation is based on two neuropsychological theories: the theory of three attentional networks (visual orienting, sustained attention, executive attention; Posner and Raichle, 1994, Fan et al., 2002) and the theory of Smith and Jonides (1999) regarding storage and executive processes, which suggests five executive components (attention and inhibition, coding, monitoring, planning, task management). Both theories rely on neuronal networks, which have been shown to be impaired in MDD.
Fig. 1 presents the linking of both theories. On the left-hand side, the three independent networks of attention are depicted; and on the right-hand side, the five components of executive functions are represented. The first component of the executive functions, namely attention and inhibition, and the executive attention network are theoretically identical and provide a link between the two theories. The five components of executive functions of Smith and Jonides (1999) are not independent. An impairment of executive attention probably has a negative influence on the performance of the four other components.
The aim of our study was to examine performance with respect to the three attentional networks and the five components of executive functions in remitted MDD patients with differing courses of illness. We predicted impaired performance in attention and executive function for MDD patients in comparison to healthy controls. In addition, we postulated a lower outcome in terms of executive functions in patients with a history of three or more depressive episodes compared to patients with a history of 1–2 depressive episodes. A matter of particular interest was the investigation of the presumed association between attentional and executive impairments in depression.
Section snippets
Subjects
The clinical sample consisted of 40 former MDD patients from the University Clinic and Policlinic of Psychiatry and Psychotherapy. Diagnoses were made using the Structured Clinical Interview (SKID, Wittchen et al., 1997) for DSM-IV Diagnoses (Saß et al., 1998). All patients were in remitted state at the time of testing. Remission was defined as a period of at least 3 months during which the criteria of a depressive episode should not be fulfilled. Patients with comorbid axis-I disorders were
Visual orienting (VAV)
In Table 2 means and standard deviations of reaction times are shown for the congruent and incongruent conditions in VAV. The mean reaction time (averaged across both conditions) of the patient groups was 70 ms slower than that of the controls (d = 0.97; t(57) = 3.55, p < 0.001, one-tailed). The congruence effect of the patient group outperformed the effect of the control group by 37 ms (d = 0.51, t(176) = 1.87, p < 0.01, one-tailed). There were no significant differences between the patient groups
Discussion
Our study expanded on the results found by Weiland-Fiedler et al. (2004) of continuing deficits in sustained attention and strategic aspects of working memory. This study investigated remitted MDD patients with mild or severe course of illness and considered the possible effects of attentional disturbance on executive functions.
We predicted impaired performance in attention and executive function for MDD patients in comparison to healthy controls. Furthermore, we postulated that MDD patients
Acknowledgement
We are indebted to the invaluable help of Prof. Dr. Med. Dr. h.c.A. Marneros, PD Dr. Med. P. Brieger and Dipl.-Psych. M. Paelecke.
References (61)
Neuroimaging studies of mood disorders
Biological Psychiatry
(2000)- et al.
Executive dysfunction predicts nonresponse to fluoxetine in major depression
Journal of Affective Disorders
(2000) - et al.
Importance of deficits in executive functions
Lancet
(1999) - et al.
Cognitive disturbances in outpatient depressed younger adult: evidence of modest impairment
Biological Psychiatry
(2001) - et al.
Neuropsychological functioning in major depression and responsiveness to selective serotonin reuptake inhibitors antidepressants
Journal of Affective Disorders
(2004) - et al.
Cognitive dysfunctions in recovered melancholic patients
Journal of Affective Disorders
(1994) - et al.
The role of executive functioning in CBT: a pilot study with anxious older adults
Behaviour Research and Therapy
(2005) - et al.
Dementia: the estimation of premorbid intelligence levels using a new adult reading test
Cortex
(1978) - et al.
Evidence for continuing neuropsychological impairments in depression
Journal of Affective Disorders
(2004) Konzentrations-Verlaufs-Test K-V-T
(1974)