Elsevier

Biological Psychiatry

Volume 49, Issue 5, 1 March 2001, Pages 416-425
Biological Psychiatry

Original article
Electroencephalographic and perceptual asymmetry differences between responders and nonresponders to an SSRI antidepressant22

https://doi.org/10.1016/S0006-3223(00)01016-7Get rights and content

Abstract

%Background:Recent reports suggest the value of electroencephalographic and dichotic listening measures as predictors of response to antidepressants. This study examines the potential of electroencephalographic alpha asymmetry and dichotic measures of perceptual asymmetry as predictors of clinical response to 12 weeks of treatment with fluoxetine (Prozac).

Methods:Resting electroencephalography (eyes open and eyes closed) and dichotic listening with word or complex tone stimuli were assessed in depressed outpatients during a pretreatment period.

Results:Fluoxetine responders (n = 34) differed from nonresponders (n = 19) in favoring left over right hemisphere processing of dichotic stimuli. They also differed in their resting electroencephalographic alpha asymmetry, particularly in the eyes open condition. Nonresponders showed an alpha asymmetry indicative of overall greater activation of the right hemisphere than the left, whereas responders did not. The relationship between hemispheric asymmetry and treatment response interacted with gender, being evident among depressed women but not men.

Conclusions:The results are consistent with the hypothesis that a characteristic tendency toward greater left than right hemisphere activation is associated with favorable response to fluoxetine, whereas the opposite hemispheric asymmetry predicts poor response.

Introduction

Selective serotonin reuptake inhibitors (SSRIs) are the most frequently prescribed antidepressant medications, and yet, they are effective in less than two thirds of depressed patients. The reason why they work in some patients but not in others is poorly understood, and there are no clinical predictors of whether or not patients will benefit from an SSRI. Although some studies have reported evidence of biological predictors of response to antidepressants Figueras et al 1999, Ko et al 1997, there are no established markers available for clinical use. Studies using neuroimaging Baxter et al 1989, George et al 1994, electrophysiologic measures Bruder et al 1997a, Henriques and Davidson 1991, and behavioral laterality tests (Bruder 1995) indicate that asymmetric activation of right-left brain regions plays a role in depressive disorders. There have also been recent reports suggesting the potential value of these measures as predictors of response to treatments for depression Bruder et al 1999, Ketter et al 1999.

Individual differences among depressed patients on dichotic listening tests of brain laterality and on electroencephalographic (EEG) measures have been found to be related to antidepressant response (for a review, see Bruder et al 1999). In dichotic listening tests, different words, syllables, or tones are simultaneously presented to the two ears and the difference in accuracy for reporting right and left ear items provides a measure of perceptual asymmetry (PA). In a collaborative study of the SSRI fluoxetine (Prozac), unmedicated depressed patients who subsequently responded to fluoxetine had a greater right ear (left hemisphere) advantage for dichotic words and less left ear (right hemisphere) advantage for complex tones when compared with treatment nonresponders (Bruder et al 1996). There were no changes in PAs following treatment, which suggests that these differences between fluoxetine responders and nonresponders represent stable, state-independent characteristics. These findings support the hypothesis that a characteristic tendency for relatively greater left than right hemisphere activation during dichotic listening is associated with better outcome of treatment with fluoxetine.

Electroencephalographic studies of regional hemispheric activity in depressed patients have focused on measures of alpha power because of its inverse relation to cortical activation (Shagass 1972). Studies measuring resting EEG have found greater alpha power over left frontal sites than over right in unipolar major depression Bell et al 1998, Henriques and Davidson 1991 and bipolar seasonal affective disorder (Allen et al 1993), which is indicative of relatively less left frontal activation or greater right frontal activation. There have, however, been some inconsistent findings concerning resting frontal alpha asymmetry in depression (e.g., Reid et al 1998), which may be related to methodological factors or the comorbidity and clinical heterogeneity typical of depressive disorders (Davidson 1998). Some studies have also found the opposite pattern of greater alpha power (less activation) over right parietal sites than over left in depressed patients Bruder et al 1997a, Reid et al 1998 or previously depressed subjects (Henriques and Davidson 1990), whereas other studies have not found this posterior asymmetry in depressed subjects Henriques and Davidson 1991, Schaffer et al 1983. Heller et al (1995) suggested that inconsistent findings for parietal alpha asymmetry may be due to the opposing effects of depression and anxious arousal on right parietotemporal activity. This is supported by the findings of Bruder et al (1997a), who compared the alpha asymmetry of depressed patients with or without a comorbid anxiety disorder. Patients having a “pure” major depression showed an alpha asymmetry indicative of less activation over right parietal sites than over left, whereas patients having a comorbid anxious depression showed evidence of greater activation over right frontal and parietal sites.

The above EEG studies suggest that depression is associated with reduced left frontal and right parietal activation. The presence of abnormal frontal and parietal alpha asymmetries in previously depressed patients supports the view that they represent state-independent markers of vulnerability to negative affect and depressive disorders (Henriques and Davidson 1990). There are, however, marked individual differences in alpha asymmetry among depressed patients, which appear to be related to their clinical features, including comorbidity and diagnostic subtype. In this study we examine whether or not individual differences in resting alpha asymmetry are related to outcome of treatment with an SSRI antidepressant. If the differences in PA observed between fluoxetine responders and nonresponders reflect characteristic, state-independent differences in regional hemispheric activation (Bruder et al 1996), one would predict that these subgroups would also differ in their resting alpha asymmetry patterns. Moreover, the PA scores for these patients should correlate with their alpha asymmetry (Davidson and Hugdahl 1996). To test these predictions, the depressed patients in this study were tested on both resting EEG and the dichotic listening tests used in our prior study.

A secondary purpose of this study was to begin to examine the possible role of gender in this context. The greater incidence of depression in women and gender differences in hemispheric asymmetries for cognitive processing underscore the importance of examining this variable (Heller 1993). Electroencephalographic studies of depression have primarily used female participants or have not examined gender differences in regional hemispheric asymmetries. In this study we examine whether gender is of importance when examining differences in hemispheric activation between fluoxetine responders and nonresponders.

Section snippets

Subjects

Patients between the ages of 18 and 65 who met DSM-IV criteria for major depression, dysthymia, or depression not otherwise specified were included in the study. Patients were excluded for any of the following reasons: serious suicide risk, seizure disorder, organic mental disorders, substance use disorders (including alcohol abuse) within the last 6 months, psychotic disorders, antisocial personality disorder, history of head trauma, or other neurologic disorder. In addition, patients were

Dichotic listening

The pretreatment differences in PA between fluoxetine responders and nonresponders replicate those seen in our prior study (Bruder et al 1996). There was an overall trend for responders to have a larger right ear (left hemisphere) advantage for words and a smaller left ear (right hemisphere) advantage for complex tones relative to nonresponders. The common direction of this group difference in asymmetry for the word and tone tests was reflected in a main effect of Treatment Response [F(1,47) =

Discussion

Individual differences in hemispheric asymmetries among depressed patients, as measured by dichotic listening or resting EEG, were related to therapeutic response to an SSRI antidepressant. In accordance with our prior study (Bruder et al 1996), fluoxetine responders differed from nonresponders in favoring left over right hemispheric processing of dichotic stimuli. Moreover, fluoxetine nonresponders differed from responders in showing a resting EEG alpha asymmetry indicative of overall greater

Acknowledgements

This work was supported in part by National Institute of Mental Health Project Grants Nos. MH36295 and MH56058.

The authors thank Donald F. Klein and reviewers of the draft of this article for their helpful comments; Jürgen Kayser for software development; Donald Ross for statistical advice; Deborah Deliyannides and other members of the Depression Evaluation Service, where diagnostic evaluations and treatment of patients were conducted; and Freedom From Fear for help with patient recruitment.

References (47)

  • R.T Knight et al.

    The effects of frontal and temporal-parietal lesions on the auditory evoked potentials in man

    Electroencephalogr Clin Neurophysiol

    (1980)
  • H.C Ko et al.

    Plasma free 3-methoxy-4-hydroxyphenylglycol predicts response to fluoxetine

    Biol Psychiatry

    (1997)
  • R.C Oldfield

    The assessment and analysis of handednessThe Edinburgh Inventory

    Neuropsychologia

    (1971)
  • J.J Sidtis

    The complex tone testImplications for the assessment of auditory laterality effects

    Neuropsychologia

    (1981)
  • B.E Wexler et al.

    Increasing the power of dichotic methodsThe fused rhymed words test

    Neuropsychologia

    (1983)
  • L.R Baxter et al.

    Reduction of prefrontal cortex glucose metabolism common to three types of depression

    Arch Gen Psychiatry

    (1989)
  • A.T Beck et al.

    An inventory for measuring depression

    Arch Gen Psychiatry

    (1961)
  • J.S Bendat et al.

    Random DataAnalyses and Measurement Procedures

    (1971)
  • G.E Bruder

    Cerebral laterality and psychopathologyPerceptual and event-related potential asymmetries in affective and schizophrenic disorders

  • G.E Bruder et al.

    Outcome of cognitive-behavioral therapy for depressionRelation to hemispheric dominance for verbal processing

    J Abnorm Psychol

    (1997)
  • G.E Bruder et al.

    Predictors of therapeutic response to treatments for depressionA review of electrophysiologic and dichotic listening studies

    CNS Spectrums

    (1999)
  • C.E Coffey et al.

    Regional cerebral blood flow correlates of a dichotic listening task

    J Neuropsychiatry

    (1989)
  • R.J Davidson

    Anterior electrophysiological asymmetries, emotion, and depressionConceptual and methodological conundrums

    Psychophysiology

    (1998)
  • Cited by (0)

    1

    Data from two treatment protocols were combined so as to yield sufficient samples of female and male fluoxetine responders and nonresponders. With the exception of the initial placebo period in one study, the treatment protocols were comparable in terms of both fluoxetine doses and the raters evaluating treatment response. Most importantly, the differences between fluoxetine responders and nonresponders reported for the total samples were also evident when only the data for the placebo-controlled protocol were analyzed; however, the larger total sample allowed us to also take patient gender into account.

    2

    The 18 healthy women were screened for psychopathology using a structured interview schedule and were excluded if they had a hearing loss, substance abuse, a history of head trauma, or other neurologic disorder. They did not differ significantly from the female patients in education (mean = 15.8, SD = 1.3) or handedness (LQ = 84.7, SD = 20.6), but they were somewhat younger than the female patients [mean age = 27.6, SD = 6.9;t(35) = 3.21,p = .001]. Age was not, however, related to perceptual asymmetry scores of either female patients (r = −.01, ns) or male patients (r = −.02, ns) on the fused words test. Nor was age significantly correlated with alpha asymmetry scores of female patients (r = −.09, ns) or male patients (r = .11, ns).

    View full text