Research reportExecutive dysfunction predicts nonresponse to fluoxetine in major depression
Introduction
For the past decade, studies of brain metabolism, cerebral blood flow, and brain electrical activity have consistently revealed that specific regions of the prefrontal cortex and basal ganglia, particularly on the left, are hypoactive in subjects with major depression as compared to control subjects. For example, using fluorodeoxyglucose positron emission tomography (FDG-PET), Baxter and colleagues (Baxter et al., 1985, Baxter et al., 1989), found decreases in brain metabolism in the left dorsolateral prefrontal cortex and caudate in subjects with major depression, bipolar depression, and subjects with OCD and depression versus normal controls. Studies of blood flow have yielded similar results, with most studies reporting reduced regional cerebral blood flow (RCBF) in the left dorsolateral prefrontal cortex and left anterior cingulate (Dolan et al., 1992, Dolan et al., 1994, Bench et al., 1993). Finally, electrophysiological studies have revealed frontal EEG asymmetries, with relatively less activation of the left as opposed to right frontal cortex, in depressed subjects as compared to controls (Henriques and Davidson, 1991), in euthymic subjects with a previous history of depression versus those with no such history (Henriques and Davidson, 1990), and in subjects reporting greater negative affect in response to films (Tomarken et al., 1990).
Furthermore, lesion studies have suggested that major depression is more common following left anterior cortical lesions and left basal ganglia lesions, particularly in the left head of the caudate (Robinson et al., 1984, Mendez et al., 1989) than following right hemisphere lesions, although not all studies have revealed similar findings (Ross and Rush, 1981).
The results of both functional and structural brain imaging studies have been interpreted to suggest that dysfunction along frontal-striatal-thalamic circuits is associated with depression, and may be involved in its pathogenesis. Alexander et al., 1986, discussed five parallel, closed, cortico-striatal-thalamic circuits that project from discrete areas of the frontal cortex to the striatum, globus pallidus/substantia nigra, thalamus, and back to the frontal cortex. One of these five circuits, termed the ‘dorsolateral prefrontal-subcortical circuit’ is thought to be involved in mediation of the ‘executive functions’; or those cognitive and behavioral functions involved in problem-solving and organizational strategies, shifting and maintaining cognitive and behavioral sets, and systematic memory search such as that used in word list generation and retrieval tasks (Mega and Cummings, 1994). This same circuit is thought to be involved in mood regulation through inhibitory interconnections with cortico-limbic circuits (Alexander et al., 1986).
While the neuropsychological deficits associated with depression have been well studied, there have been few attempts to link neuropsychological data with results of brain imaging studies into a coherent theoretical framework. In addition, generalizing about the neuropsychology of depression is difficult because studies vary considerably in terms of sample composition and measures used to assess cognition. Nevertheless, results of well-controlled studies reveal that, as a group, depressives perform significantly worse than control subjects on measures of speed-of-information processing, motor speed, verbal and nonverbal learning but not retention or recognition memory (and may perform particularly poorly on nonverbal memory tests), and tests of executive functions, particularly those that measure cognitive flexibility, problem-solving, semantic retrieval, working memory, and response inhibition (Jones et al., 1988, Martin et al., 1991, Otto et al., 1994, Norris et al., 1995, Boone et al., 1995, Trichard et al., 1995, Channon, 1996, Elliott et al., 1996, Purcell et al., 1997, Marshall et al., 1997, Merriam et al., 1999).
It has been hypothesized that the neuropsychological profile of depression is more similar to profiles of patients with disease states that involve the frontal lobes and/or subcortical structures, such as patients with frontal and/or subcortical infarctions from cerebrovascular disease, Parkinson’s disease, or Huntington’s disease, rather than patients with primary cortical degeneration, such as Alzheimer’s disease (Massman et al., 1992). For example, neuropsychological test scores were used to compare the performance of patients with unipolar and bipolar depression to the performance of performance of patients with subcortical disease (Huntington’s disease), primary cortical degeneration (Alzheimer’s disease), and normal controls using discriminant function analysis (Massman et al., 1992). While almost half (49%) of the depressed subjects were classified as normals, a sizable minority (28.6%) were classified as Huntington’s disease. These patients were not distinguishable from the depressed subjects performing similar to control subjects on the basis of education, age, psychiatric history, or severity of depression. However, other studies (Tham et al., 1997, Basso and Bornstein, 1997) have found that those subjects displaying cognitive impairment have had a greater number of past episodes of depression than those without impairment. Therefore, the reasons for this heterogeneity in cognitive functioning are not well understood.
In this study, we sought to explore the heterogeneity in cognition among depressed patients further by investigating the relationship between pretreatment neurocognitive performance and treatment outcome. Few consistent predictors of response to antidepressant medications have been identified, although there are some preliminary data from our laboratory (Cook et al., 1999) and others (Knott et al., 1996, Mayberg et al., 1997) suggesting that subtle, baseline brain dysfunction is predictive of poor response to medication. To our knowledge, the relationship of cognitive functioning to treatment outcome has not been widely explored in the literature, although one recent study found that executive dysfunction on cognitive tasks as well as delayed P300 auditory evoked potential latency was predictive of outcome in geriatric depressed subjects (Kalayam and Alexopoulos, 1999). We hypothesized that depressed subjects with poor treatment response would display greater pretreatment cognitive deficits than subjects with good treatment response. We investigated this question in the context of a randomized, double-blind, placebo-controlled clinical trial.
Section snippets
Subjects
The sample consisted of 14 individuals diagnosed with Major Depressive Episode using DSM-IV criteria and the Structured Clinical Interview for DSM-IV (SCID-P; First et al., 1994) who participated in a randomized, double-blind, placebo-controlled trial of the effectiveness of fluoxetine in treating major depression. Only subjects assigned to active treatment (versus placebo) are included in the present report. Subjects were required to score 18 or above on the 17-item version of the Hamilton
Demographic variables and rating scales
Means and standard deviations of demographic variables and rating scale scores are presented in Table 1, while Table 2 presents the demographics and rating scale scores broken down by clinical outcome group. Using the criterion outlined above to determine clinical outcome group (responders must have a final HAM-D score of less than or equal to 10), 14 subjects were classified as responders, while eight were classified as nonresponders. In order to investigate whether any individual nonresponder
Discussion
Our findings suggest that lower pretreatment scores on measures of executive functioning are predictive of nonresponse to fluoxetine in subjects with major depression. The differences between responders and nonresponders on these measures does not appear to be solely a function of age, past history of depression, education, estimated premorbid IQ, or severity of depression, given that the variance attributed to baseline group differences in depression was statistically controlled, and the
Acknowledgements
This work was supported by grants R01 MH40705 and Research Scientist Development Award K02-MHO1165 from the National Institutes of Mental Health to Dr. Leuchter, a Young Investigator Award from the National Alliance for Research in Schizophrenia and Depression (NARSAD) to Dr. Cook, and a grant from Eli Lilly and Company, Inc. The authors would like to thank three anonymous reviewers for their comments on an earlier draft of this manuscript. Portions of this manuscript were presented at the 26th
References (52)
Executive dysfunction in depression: The Wisconsin Card Sorting Test
J Affect Disord
(1996)- et al.
Neurophysiologic predictors of treatment response to fluoxetine in major depression
Psychiatry Res
(1999) Gender differences in depression: perspectives from neuropsychology
J Affect Disord
(1993)- et al.
Quantitative EEG in the prediction of antidepressant treatment response to imipramine
J Affect Disord
(1996) - et al.
Norms for depressed patients for the California Verbal Learning Test: Associations with depression severity and self-report of cognitive difficulties
Arch Clin Neuropsychol
(1994) - et al.
The epidemiology of depression: An update on sex differences in rates
J Affect Disord
(1984) - et al.
Parallel organization of functionally segregated circuits linking basal ganglia and cortex
Ann Rev Neurosci
(1986) - et al.
T2 hyperintensities in bipolar disorder: magnetic resonance imaging comparison and literature meta-analysis
Am J Psychiatry
(1995) - et al.
Relative memory deficits in recurrent versus first- episode major depression
- et al.
Cerebral metabolic rates for glucose in mood disorders: Studies with positron emission tomography and fluorodeoxyglucose F18
Arch Gen Psychiatry
(1985)