Amygdala and insula volumes prior to illness onset in bipolar disorder: A magnetic resonance imaging study

https://doi.org/10.1016/j.pscychresns.2011.06.010Get rights and content

Abstract

There are now numerous reports of neuroanatomical abnormalities in people with bipolar disorder. However, it remains unclear whether those abnormalities predate the onset of the illness. In this cross-sectional magnetic resonance imaging study, we assessed 11 young people clinically at ultra-high risk of development of psychosis (UHR), who all developed bipolar I or II disorder by follow-up (median time to onset 328 days – UHR-BP), 11 matched UHR participants, who had no psychiatric diagnosis after at least 12 months of follow-up (UHR-Well) and 11 matched healthy controls (HC). Our main outcome measures were amygdala, hippocampus, insula, lateral ventricular and whole brain volumes. Amygdala and insula volume reductions were more pronounced in the UHR-BP than in the UHR-Well and HC group. Lateral ventricle, whole-brain and hippocampal volumes did not differ between groups. If these findings are confirmed, they suggest that imaging investigations could help to distinguish people who will subsequently develop bipolar disorder from those who will not, at least in symptomatically enriched samples.

Introduction

While disease-related neuroanatomical abnormalities have been identified in bipolar disorder (Soares and Mann, 1997, Beyer and Krishnan, 2002, McDonald et al., 2004, Kempton et al., 2008), it remains unclear whether these abnormalities predate the onset of the illness. The most rigorous approach to determining whether neuroanatomical changes predate onset is to recruit individuals manifesting non-specific at-risk features prior to the onset of threshold disorder; and prospectively monitor them, enabling ascertainment of diagnostic outcomes and comparison of baseline and longitudinal changes between those who do and do not make the transition to frank illness. We have adopted such an approach to the study of psychosis for the past 15 years (Yung et al., 1996, Yung et al., 2003). Although the approach is specifically aimed at the prediction of psychotic illnesses, a small number of these at-risk individuals progressed to threshold bipolar disorder (BD). Although the findings in this population might not be generalizable to all people with BD, this sample offers the unique opportunity to compare neuroanatomical measures prior to the onset of first episode mania with those of individuals who meet ultra-high risk criteria, but did not meet diagnostic criteria for a psychiatric disorder at follow-up, and with data from healthy comparison subjects. An added advantage of our cohort is that the participants are free of mood stabilizers, which is an important issue given that lithium has been found to increase gray matter volume in vivo (Moore et al., 2000, Monkul et al., 2007, Phillips et al., 2008).

Predicting exactly where anatomical differences might be seen prior to the onset of BD is difficult, since the neurobiology of the illness is not well understood. Although published data are inconsistent, they do suggest disease-related neuroanatomical abnormalities in the prefrontal cortex, as well as the subcortical and medial temporal structures (Soares and Mann, 1997, Beyer and Krishnan, 2002, McDonald et al., 2004, Kempton et al., 2008, Yeh et al., 2010). These inconsistencies may at least partially reflect differences in age and chronicity of patient groups or medication effects. For example, decreased amygdala volumes have been reported in adolescents with BD (DelBello et al., 2004, Hajek et al., 2009b), whereas amygdala enlargement has been reported in several studies of adult patients (Altshuler et al., 2000, Brambilla et al., 2003, Hajek et al., 2009b, Usher et al., 2010) as well as in patients who have been treated with mood stabilizers (Savitz et al., 2010). In addition, multiple-episode patients have been shown to have significantly larger lateral ventricles than first-episode patients (Brambilla et al., 2001, Strakowski et al., 2002). Family history may play an important role, as abnormalities in certain brain structures have been more frequent in familial patients (e.g., Drevets et al., 1998). However, two recent studies found that the offspring of individuals with bipolar I disorder did not exhibit structural abnormalities in regions that are found to be abnormal in adolescents and adults already diagnosed with bipolar disorder (Singh et al., 2008, Hajek et al., 2009a). Moreover, nine male children, who initially presented with psychosis not otherwise specified, attention deficit hyperactivity disorder and mood liability combined, and who developed bipolar disorder at 2- to 4- year follow-up, did not differ from controls regarding cortical gray matter (Gogtay et al., 2007). Therefore, given this heterogeneity and the novelty of our cohort, we did not make specific hypotheses but chose to investigate a number of key brain regions implicated in BD, including the amygdala, hippocampus, and insula, as well as lateral ventricular and whole brain volumes (Hajek et al., 2005).

Section snippets

Setting and participants

We recruited people at ultra-high risk (UHR) of developing a psychotic disorder from the Personal Assessment and Crisis Evaluation (PACE) Clinic in Melbourne, Australia (McGorry et al., 2001a, McGorry et al., 2001b, Yung et al., 2003). Criteria for the identification of the UHR cohort and the rationale for these criteria have been previously described (Yung et al., 2005). The aim of this approach was to identify help-seeking, troubled young people who have manifest symptoms and impaired

Demographics and clinical characteristics

Demographic details and clinical characteristics are presented in Table 2. The Global Assessment of Functioning (GAF) score was significantly higher in the UHR-Well group at intake (t(19) = 2.16, p = 0.044). There were no significant differences between the groups on any other measure, including the baseline HAM-D (t(12.7) = 1.53, p = 0.15) and YMRS (t(19) = 0.96, p = 0.35).

Intracranial and whole-brain volume

Intracranial volume was compared across the three groups (HC, UHR-BP and UHR-Well) using an ANCOVA that controlled for height. There

Discussion

This is the first study to neuroanatomically investigate adolescents and young adults with BD prior to the onset of first episode mania. The main findings are that amygdala and insula volume reductions are present prior to the onset of first episode mania, whereas intracranial, lateral ventricle, whole-brain and hippocampal volumes are not different from a UHR well or a healthy comparison group. These abnormalities cannot be ascribed to the effects of mood stabilizers or antipsychotic drugs,

Acknowledgments

This study was supported by project grants from the National Health & Medical Research Council (NHMRC; grant IDs: 145627, 145737, 970598, 981112, 970391), NHMRC Program Grant (ID: 350241), and the Colonial Foundation. Dr Bechdolf was supported by a grant from the German Research Foundation (Be 3697/1-1). Prof. Wood was supported by a Clinical Career Development Award from the NHMRC (ID: 359223). Dr. Takahashi was supported to undertake this work by a Grant-in-Aid for Scientific Research (No.

References (52)

  • C.D. Ladouceur et al.

    Subcortical gray matter volume abnormalities in healthy bipolar offspring: potential neuroanatomical risk marker for bipolar disorder?

    Journal of the American Academy of Child and Adolescent Psychiatry

    (2008)
  • J. LeDoux

    The amygdala

    Current Biology

    (2007)
  • C. McDonald et al.

    Meta-analysis of magnetic resonance imaging brain morphometry studies in bipolar disorder

    Biological Psychiatry

    (2004)
  • M.P. Milham et al.

    Selective reduction in amygdala volume in pediatric anxiety disorders: a voxel-based morphometry investigation

    Biological Psychiatry

    (2005)
  • E.S. Monkul et al.

    Prefrontal gray matter increases in healthy individuals after lithium treatment: a voxel-based morphometry study

    Neuroscience Letters

    (2007)
  • G.J. Moore et al.

    Lithium-induced increase in human brain grey matter

    The Lancet

    (2000)
  • J.C. Pfeifer et al.

    Meta-analysis of amygdala volumes in children and adolescents with bipolar disorder

    Journal of the American Academy of Child and Adolescent Psychiatry

    (2008)
  • J. Savitz et al.

    Amygdala volume in depressed patients with bipolar disorder assessed using high resolution 3 T MRI: the impact of medication

    Neuroimage

    (2010)
  • M.K. Singh et al.

    Neuroanatomical characterization of child offspring of bipolar parents

    Journal of the American Academy of Child and Adolescent Psychiatry

    (2008)
  • J.C. Soares et al.

    The anatomy of mood disorders—review of structural neuroimaging studies

    Biological Psychiatry

    (1997)
  • T. Takahashi et al.

    Follow-up MRI study of the insular cortex in first-episode psychosis and chronic schizophrenia

    Schizophrenia Research

    (2009)
  • J. Usher et al.

    Correlation between amygdala volume and age in bipolar disorder—a systematic review and meta-analysis of structural MRI studies

    Psychiatry Research

    (2010)
  • P.H. Yeh et al.

    Structural equation modeling and principal component analysis of gray matter volumes in major depressive and bipolar disorders: differences in latent volumetric structure

    Psychiatry Research

    (2010)
  • A.R. Yung et al.

    Psychosis prediction: 12 month follow-up of a high risk ('prodromal') group

    Schizophrenia Research

    (2003)
  • N.C. Andreasen

    The scale for the assessment of negative symptoms (SANS)

    (1983)
  • J.L. Beyer et al.

    Volumetric brain imaging findings in mood disorders

    Bipolar Disorders

    (2002)
  • Cited by (45)

    • Grey and white matter structure associates with the activation of the tryptophan to kynurenine pathway in bipolar disorder

      2019, Journal of Affective Disorders
      Citation Excerpt :

      The amygdala and the insula are involved in evaluative and affective processes, with the amygdala being associated with affective arousal, in particular negative stimuli, while the insula integrate affective and cognitive processes (Berntson et al., 2011). A reduced GM volume in the amygdala and insula have been shown in ultra-high risk individuals of developing BD (Bechdolf et al., 2012). In addition, lower amygdala-insula connectivity for the threat emotions and greater amygdala-insula connectivity for the sad emotions have been shown in BD compared with major depressive disorder (Korgaonkar et al., 2019).

    • Polygenic Score of Subjective Well-Being Is Associated with the Brain Morphology in Superior Temporal Gyrus and Insula

      2019, Neuroscience
      Citation Excerpt :

      There were also many studies that indirectly explored the relationship between the two. For example, research on depression linked it to SWB (Beekman et al., 1997; Wood and Joseph, 2010; Okbay et al., 2016) and depressed individuals showed a decrease of GMV of the right insula (Hwang et al., 2010; Reiner et al., 2011; Andreas et al., 2012). Some researchers also found the insula was involved in social cognition and emotional processing (Uddin et al., 2017; Caldiroli et al., 2018).

    • Amygdala Activation and Connectivity to Emotional Processing Distinguishes Asymptomatic Patients With Bipolar Disorders and Unipolar Depression

      2019, Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
      Citation Excerpt :

      Support for this interpretation can be found in the work of Ellard et al. (35), who also found differences in resting-state functional connectivity between the insula and the inferior parietal node of the frontoparietal executive brain network to distinguish unipolar and bipolar depressed participants and other reports of differential resting connectivity related to both the amygdala and the insula (36,37). Further, reduced insula and amygdala volumes in individuals at ultra-high risk for the development of psychosis who then subsequently transition to BP, as compared with those who do not, lends further support to the involvement of both of these regions as risk trait markers for BP (38). Consistent with earlier observations in depressed versus remitted BP cohorts (14), we also found reduced connectivity between the amygdala and mOFC during supraliminal happy stimuli for BP relative to MDD, which was characteristic in differentiating participants with remitted MDD but not participants with remitted BP from HC subjects.

    View all citing articles on Scopus
    View full text