Regional cerebral glucose metabolism and anxiety symptoms in bipolar depression: Effects of levothyroxine

Psychiatry Research: Neuroimaging 181 (2010) 71–76

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s n s

Regional cerebral glucose metabolism and anxiety symptoms in bipolar depression: Effects of levothyroxine Michael Bauer a,b,⁎, Steven M. Bermanb, Florian Schlagenhauf c, Bradley Voytekb, Natalie Rasgond, Mark A. Mandelkerne, Peter C. Whybrow b, Edythe D. Londonb a

Department of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA, USA c Department of Psychiatry and Psychotherapy, Charité — Universitätsmedizin Berlin, Campus Mitte, Berlin, Germany d Department of Psychiatry and Behavioral Sciences, Stanford School of Medicine, Palo Alto, CA, USA e Department of Physics, University of California Irvine, Irvine, CA, USA b

a r t i c l e

i n f o

Article history: Received 28 April 2008 Received in revised form 8 April 2009 Accepted 5 July 2009 Keywords: Thyroid hormone Levothyroxine Bipolar disorder Anxiety Positron emission tomography Depression

a b s t r a c t We examined the relationships between regional brain activity and anxiety in bipolar depressed patients receiving adjunctive treatment with levothyroxine. Regional brain activity was assessed with positron emission tomography and [18F]fluorodeoxyglucose in 10 euthyroid, depressed bipolar women before and after 7 weeks of adjunctive therapy with levothyroxine. The primary biological measures were relative (to global) regional radioactivity as a surrogate index of glucose metabolism in pre-selected brain regions. Relationships were assessed between regional brain activity and anxiety symptoms while controlling for depression severity. At baseline, Trait Anxiety Inventory measures covaried positively with relative brain activity bilaterally in the dorsal anterior cingulate, superior temporal gyri, parahippocampal gyri, amygdala, hippocampus, ventral striatum, and right insula; state anxiety showed a similar pattern. After treatment anxiety was improved significantly. Change in trait anxiety covaried positively with changes in relative activity in right amygdala and hippocampus. Change in state anxiety covaried positively with changes in relative activity in the hippocampus bilaterally and left thalamus, and negatively with changes in left middle frontal gyrus and right dorsal anterior cingulate. Results indicate that comorbid anxiety symptoms have specific regional cerebral metabolic correlates in bipolar depression and cannot only be explained exclusively by the depressive state of the patients. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Epidemiological studies have widely documented a high rate of comorbidity between bipolar and anxiety disorders (Wittchen et al., 1994). An independent association of comorbid anxiety with greater severity and impairment in bipolar disorder patients (Simon et al., 2004) and an association of high anxiety levels with poor therapeutic outcome have also been demonstrated (Frank et al., 2002). Although this relationship is well established, the underlying neurobiology of anxiety in bipolar disorder has not been thoroughly studied (Freeman et al., 2002). Functional neuroimaging studies in patients with primary anxiety disorders have provided evidence of higher rates of regional cerebral glucose metabolism, particularly in hippocampal and parahippocampal areas, compared to controls (Bisaga et al., 1998; Rauch et al., 2003). ⁎ Corresponding author. Department of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität Dresden, Fetscherstr. 74, 01307 Dresden, Germany. Tel.: +49 351 458 2772; fax: +49 351 458 4324. E-mail address: [email protected] (M. Bauer). 0925-4927/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2009.07.001

Other neuroimaging studies have identified the orbitofrontal, insular, temporal, cingulate, parietal and occipital cortices as important neural substrates of anxiety disorders (Rauch et al., 1997). In a recent positron emission tomography (PET) study, Sakai et al. (2005) showed higher [18F]fluorodeoxyglucose (FDG) uptake in the amygdala and hippocampus in patients with panic disorder experiencing high state and trait anxiety before entering the scanner compared to healthy controls. Abnormally elevated indices of glucose metabolism have also been observed in limbic and subcortical (striatum and thalamus) structures of patients with unipolar depression (Drevets, 2000) and bipolar depression (Ketter et al., 2001; Strakowski et al., 2005). Studies of bipolar disorder, however have rarely addressed the relationship between anxiety symptoms and regional brain function. One study demonstrated that comorbid anxiety symptoms in bipolar disorder are associated with specific neural correlates (Osuch et al., 2000). In that study, after covariation for depression scores, severity of anxiety correlated positively with regional glucose metabolism in right parahippocampal and left anterior cingulate regions, and inversely with glucose metabolism in the cerebellum, left fusiform gyrus, superior temporal gyrus, angular gyrus, and insula.

72

M. Bauer et al. / Psychiatry Research: Neuroimaging 181 (2010) 71–76

We previously reported that augmentation treatment with levothyroxine (L-T4) improves mood (Bauer et al., 1998) and normalizes elevated relative cerebral glucose metabolism in the subgenual cingulate, amygdala, hippocampus and subcortical brain areas (thalamus, dorsal and ventral striatum, and cerebellar vermis) in women with bipolar depression (Bauer et al., 2005). The present study represents a secondary analysis of data from the same study, which used FDG and PET, and examines the relationship between parallel changes in anxiety symptoms and in relative regional activity among bipolar depressed patients receiving L-T4 treatment. We hypothesized that in bipolar depression relative brain activity in the amygdala and its related limbic structures would be positively associated with severity of anxiety.

and physical examination, a routine laboratory evaluation (thyroid function tests, blood count, blood chemistry, and urine drug screen), examination of vital signs, and a 12-lead electrocardiogram (ECG). Anxiety was self-reported using the Spielberger State and Trait Anxiety Inventory (STAI). Participants completed the STAI-State form (STAI-S; possible total scores: 0–80) and the STAI-Trait form (STAI-T; possible total scores: 0–80) immediately after PET scanning was completed. State anxiety is supposed to measure the intensity of feelings at a particular moment (anxiety-in-progress), and trait anxiety is conceptualized as the propensity of a person to experience state anxiety and is considered to be more stable and reliable (Endler and Kocovski, 2001). 2.3. Procedures

2. Methods and materials

This was a prospective, single-blind, 7-week study that investigated the efficacy of augmentation treatment with supraphysiological doses of L-T4 and the relation of clinical response to alterations in regional brain function in women with bipolar depression (Bauer et al., 2005). Only women were studied because previous work had indicated that women benefit more than do men from thyroid hormone supplementation (Bauer et al., 2008). For the purpose of this secondary analysis, anxiety symptoms were analyzed for associations with regional cerebral activity while controlling for depression severity both before and after the course of L-T4 treatment. The measure of brain function was normalized, decay-corrected, raw counts (of the radiotracer FDG) as a surrogate index of glucose metabolism. Relative activity (as used in this report) refers to this measure.

At study visits, a clinical evaluation, including assessment of adverse event, body weight, vital signs, cardiac (including ECG) and thyroid function, was performed. Levothyroxine (Levothroid®) was administered once daily in addition to the antidepressant and mood stabilizer medication that each patient was receiving but to which he/she was not responding. The L-T4 dose was 100 μg/day for the first week, 200 μg/day for the second week and 300 μg/day for weeks 3–7; if TSH was not suppressed by the end of week 3, the L-T4 dose was increased to 400 μg/ day. At study entry, patients were receiving antidepressants (mostly SSRIs and venlafaxine) and mood stabilizers (mostly lithium and divalproex sodium) (for details see Bauer et al., 2005). There was no change in treatment with any of these medications throughout the study. None of the research participants received hormone replacement therapy; all were pre-menopausal (evidenced by urine ovulation test) except for one (aged 53 years) who had a hysterectomy.

2.2. Inclusion and exclusion criteria and participants

2.4. MRI and PET imaging

Study inclusion required participants to be between the ages of 18 and 55 years, and to have a diagnosis of bipolar I or II disorder. The screening for positive bipolar disorder included a current depressive episode and a score ≥15 on the clinician-rated 21-item Hamilton Rating Scale for Depression (HRSD21) despite antidepressant therapy (≥6 weeks, standard doses of at least one antidepressant; see below). Patients had received the same antidepressant during the 6 weeks before study entry, with no changes in dose during the 3 weeks before entry. They also were euthyroid (serum TSH [thyroid-stimulating hormone] 0.3–4.7 mcIU/ml). Participants were excluded from the study if they demonstrated any of the following: psychotic features or history of schizoaffective disorder/schizophrenia; organic brain disorder; current alcohol dependence or abuse; current dependence, or abuse with a positive urine screen on an addictive substance (illegal drugs); thyroid adenoma or hypothyroid condition; unstable medical illness; endocrine disorder; severe cardiovascular disease; intake of thyroid hormone during the 4 months immediately preceding study entry; clinical judgment of serious suicidal tendency; pregnancy, lactation, or childbirth within a year prior to study entry; or childbearing potential without use of contraception, due to the susceptibility of a fetus or nursing infant to medical harm from the injected radiotracer. Further details of recruitment and diagnostic procedures of the UCLA Institutional Review Board-approved study were previously described in detail (Bauer et al., 2005). Of the 10 participants (mean age 39.3 ± 7.8 years, all of Caucasian descent), nine were diagnosed with bipolar I disorder, and one with bipolar II disorder (Structured Clinical Interview for DSM-IV Axis I Disorders). The mean duration of the current episode of depression at study entry was 171 ± 125 days. These 10 participants had 15.2 ± 2.4 years of education and were right handed (as determined by the Edinburgh Handedness Inventory). For screening and diagnostic purposes, participants completed a medical

Each participant completed two PET scanning sessions with FDG to assess relative activity, one before L-T4 treatment and again after 7 weeks of treatment. Thyroid status and psychiatric ratings were assessed on the morning of each PET scan. Beginning before injection of the FDG and continuing during the uptake period, the participant performed an auditory continuous performance task (CPT) that was administered to provide a consistent cognitive set across all participants. PET images were acquired with a Siemens ECAT EXACT HR+ tomograph (CTI, Knoxville, TN) in 63 planes with a 15.5-cm field of view (FOV) in 3D mode, as described previously (Bauer et al., 2005). After completion of a transmission scan (for attenuation correction), the participant was removed from the gantry and seated to perform the CPT. Approximately 5 min after the CPT was started, FDG (b5 mCi, b185 MBq) was administered as an intravenous bolus. Brain images were acquired for 30 min (six 5-min frames), beginning 50 min after FDG injection. T1 volumetric structural magnetic resonance imaging (MRI) scans (3-T, General Electric) were used for co-registration with PET data.

2.1. Study design

2.5. Data analysis Clinical anxiety measures before and after treatment with L-T4 were compared using Student's t tests. Statistical significance for all analyses was set at α = 0.05 (all tests two-tailed). As a surrogate index for relative brain glucose metabolism, we used decay-corrected, raw counts of radioactivity, scaled to the global mean of each scan, and made comparisons between time of assay (pre- and post-treatment) using Statistical Parametric Mapping (SPM99) (Wellcome Department of Cognitive Neurology, 2000). Each PET image was co-registered to the corresponding MRI using Automated Image Registration (Woods et al., 1993). The MR images were then used to normalize each participant's PET data spatially through linear and

M. Bauer et al. / Psychiatry Research: Neuroimaging 181 (2010) 71–76

nonlinear transformations that warped the images to a standard coordinate system developed at the Montreal Neurological Institute (MNI space). On the basis of previous studies in bipolar depression (Baxter et al., 1989; Drevets et al., 1997; Drevets, 2000; Osuch et al., 2000; Strakowski et al., 2005), we identified brain regions where we expected associations between anxiety symptoms and relative brain activity. Accordingly we sampled nine brain regions bilaterally and the cerebellar vermis at midline, producing measurements in 19 volumes of interest (VOIs). The regions sampled bilaterally were: middle frontal gyrus (selected as representative of the dorsolateral prefrontal cortex — Brodmann areas (BA) 6, 8, 9, 10, 46), dorsal anterior cingulate, superior temporal gyrus, insula, parahippocampal gyrus, amygdala, hippocampus, ventral striatum, and thalamus. The VOIs were drawn on the structural MRI template provided in SPM99, using MEDx software (Sensor Systems, Sterling, VA) with reference to the cited papers that argued for including these regions, and aided by use of the Talairach and Tourneoux (1988) atlas and the atlas of Duvernoy (1999).

73

We used the SPM99 small volume correction (SVC) approach to assess each experimental question for each VOI. The voxel height threshold for inclusion in clusters was P = 0.05 (uncorrected). A VOI was considered to show a significant treatment effect if it contained a cluster with P b 0.05 for spatial extent (corrected). For each VOI that showed a significant effect, we present the cluster size and corresponding corrected P value (Tables 1 and 2). We also noted the coordinates and P value for the peak voxel (corrected for VOI search volume) to facilitate comparisons with other studies, as described before (Bauer et al., 2005). The relationship between relative glucose metabolism and anxiety measures before treatment was tested using a SPM covariate only design. A second analysis assessed the degree to which change in relative activity after treatment was correlated with change in anxiety (STAI-T and STAI-S). Before treatment, both trait and state anxiety measures were correlated with the HRSD21 total score (STAI-T: r = 0.62; P = 0.059; STAI-S: r = 0.64; P = 0.047). After treatment, only state anxiety retained a significant correlation with the HRSD21 total score (STAI-S: r = 0.74; P = 0.015; STAI-T: r = 0.38; P = 0.284). To remove variance components related to depressive symptoms the

Table 1 Associations of anxiety measures with relative brain activity in bipolar depression at baseline (pre-treatment).

STAI-Trait positive covariate Dorsal anterior cingulate Left Right Superior temporal gyrus Left Right Insula Right Parahippocampal gyrus Left Right Amygdala Left Right Hippocampus Left Right Ventral striatum Left Right

Cluster-level analysis

Peak voxel

Corrected P value

Z-score

Per cluster

Corrected P value

0.000a 0.007

200 100

447 456

0.002 0.288

4.54 2.24

−6 2

12 22

32 26

0.050 0.000a

108 422

2665 2436

0.338 0.330

3.07 3.04

− 48 54

2 − 10

−8 −6

0.011

No. of voxels Search volume

Coordinates x

y

z

196

1456

0.327

2.73

46

−8

0

a

0.000 0.000a

141 59

241 321

0.065 0.108

2.97 2.85

− 26 32

− 14 − 18

− 30 − 28

0.043 0.001a

20 89

160 161

0.035 0.042

2.96 2.89

− 16 30

−4 2

− 14 − 14

0.003 0.000a

137 204

558 541

0.197 0.104

2.62 2.96

− 18 30

− 12 −8

− 22 − 16

0.000a 0.006

100 43

208 184

0.026 0.129

3.23 2.37

− 26 24

6 8

− 10 −8

0.021

53

447

0.114

2.82

− 10

8

38

0.006 0.003

201 228

2665 2436

0.231 0.126

3.27 3.52

− 60 52

− 22 −8

10 4

0.045 0.009

121 213

1535 1456

0.061 0.137

3.63 3.23

− 40 50

−2 −6

−6 0

0.022 0.000a

11 62

241 321

0.409 0.160

1.92 2.64

− 20 26

− 14 − 22

− 28 − 26

0.006

43

161

0.096

2.46

28

−2

− 16

0.025 0.045

65 91

541 720

0.382 0.220

2.09 2.58

26 6

−6 − 66

− 16 − 42

STAI-Trait negative covariate No significant associations STAI-State positive covariate Dorsal anterior cingulate Left Superior temporal gyrus Left Right Insula Left Right Parahippocampal gyrus Left Right Amygdala Right Hippocampus Right Cerebellar vermis STAI-State negative covariate No significant associations Relationships between self-reports of anxiety (STAI-Trait, STAI-State) and relative activity were assessed by testing whether STAI scores were a significant covariate of brain activity in nine bilateral brain regions plus the cerebellar vermis, which were selected a priori on the basis of literature accounts of involvement in anxiety states (see Section 2). a Statistically significant after Bonferroni correction for the number of comparisons (19 tests).

74

M. Bauer et al. / Psychiatry Research: Neuroimaging 181 (2010) 71–76

Table 2 Relationship between changes in anxiety (STAI) and changes of relative brain activity after L-T4 treatment in subjects with bipolar depression.

STAI-Trait positive covariate Amygdala Right Hippocampus Right

Cluster-level analysis

Peak voxel

Corrected P value

Corrected P value

Z-score

No. of voxels Per cluster

Search volume

Coordinates x

y

z

0.002a

75

161

0.059

2.60

28

−4

− 14

a

219

541

0.090

2.88

36

− 20

− 10

0.001

STAI-Trait negative covariate No significant associations STAI-State positive covariate Hippocampus Left Right Thalamus Left

0.038 0.045

61 55

558 541

0.026 0.238

3.46 2.30

− 24 26

− 32 − 20

−6 − 14

0.010

349

1392

0.154

2.92

− 18

−6

4

STAI-State negative covariate Middle frontal gyrus Left Dorsal anterior cingulate Right

0.042

115

1564

0.114

2.90

− 24

52

38

203

456

0.076

2.88

6

14

26

0.001

a

a

Statistically significant after Bonferroni correction for the number of comparisons (19 tests).

HRSD21 total score was included as a nuisance variable in all reported SPM analyses. Therefore we specifically assessed the association of trait and state anxiety with relative glucose metabolism, while controlling for depression severity. Further, we noted which effects maintained statistical significance after applying the Bonferroni correction for the number of comparisons (adjusted alpha level: 0.05/19 = 0.0026). Nonetheless, the Bonferroni method may be overly conservative as it assumes independence across tests while brain activities in the various regions sampled are not likely to be independent. 3. Results All 10 of the bipolar patients who received the L-T4 treatment completed the study; the mean L-T4 dose at the end of the study was 320 ± 42.1 μg/day (range = 300–400). L-T4 treatment caused thyroid hormone measures to increase significantly (free T4 index [reference range 4.5–10.5]: from 7.3 ± 1.7 to 19.3 ± 7.5, P b 0.01; T4 total [reference range 4.9–11.4 mcg/dl]: 7.9 ± 1.8 to 14.5 ± 3.9 mcg/dl, P b 0.001) and basal TSH levels (reference range 0.3–4.7 mcIU/ml) to decrease

significantly (from 2.2 ± 2.1 to 0.02 ± 0.0 mcIU/ml, P b 0.05). Treatment caused systolic blood pressure to decrease significantly while heart rate, diastolic blood pressure, and body weight did not significantly change (for details see Bauer et al., 2005). At baseline, trait anxiety symptoms showed a significantly positive covariation with relative brain activity bilaterally in the dorsal anterior cingulate, superior temporal gyri, parahippocampal gyri, amygdala, hippocampus, ventral striatum and right insula (Table 1). After Bonferroni correction for the number of tests (19 tests) the covariation between trait anxiety and bilateral parahippocampal gyri, right superior temporal gyri, right hippocampus and right amygdala, left dorsal anterior cingulate and left ventral striatum retained significance. STAI-S and STAI-T scales were highly inter-correlated (r = 0.75; P = 0.013). Accordingly, state anxiety showed a similar pattern of covariation with regional brain glucose metabolism at baseline (Table 1). During L-T4 treatment, patients exhibited significant declines in state (STAI-S: 53.1 ± 10.0 vs. 43.6 ± 14.7; P b 0.05) and trait anxiety scores (STAI-T: 64.6 ± 6.0 vs. 48.8 ± 14.2; P b 0.01). Changes in relative activity during L-T4 treatment were associated with measures of

Fig. 1. Brain areas where changes in regional activity were positively correlated with changes in trait anxiety after treatment with levothyroxine in bipolar depression. Statistical parametric maps were generated using SPM99. Colors superimposed on the gray-scale structural MR template indicate areas where the height threshold for the contrast (wholebrain) was t ≥ 1.69 (P = 0.05). Arrows indicate locations where clusters exhibited P b 0.05 for spatial extent (corrected for search volume of the relevant VOI but not the number of regions). Coordinates are in MNI space.

M. Bauer et al. / Psychiatry Research: Neuroimaging 181 (2010) 71–76

anxiety. Decreases in relative activity in the right amygdala and right hippocampus (significantly after Bonferroni correction) covaried positively with the reduction in trait anxiety (Table 2; Fig. 1). Change in state anxiety displayed positive covariation with change in relative activity in bilateral hippocampus and left thalamus; there was a negative covariation with right dorsal anterior cingulate (significant after Bonferroni correction) and left middle frontal gyrus (Table 2). 4. Discussion This study confirmed the intimate association between bipolar depression and moderate to severe symptoms of anxiety. The scores of anxiety reported here were comparable to studies of patients with anxiety disorders, e.g., panic disorder “at rest” (Bisaga et al., 1998) or social phobia (Tillfors et al., 2001). When patients were in the depressed state (pre-treatment), anxiety symptoms covaried positively with relative activity within subcortical limbic structures, including the amygdala, hippocampus, parahippocampal gyrus and the ventral striatum, indicating that high anxiety scores are associated with high glucose metabolism. Because we removed the variance related to depressive symptoms (as measured with the Hamilton Rating Scale for Depression) our finding of an association between anxiety symptoms and regional brain metabolism is specific and cannot only be explained exclusively by the depressive state of the patients. As reported previously (Bauer et al., 2005), some of these limbic structures (hippocampus, amygdala and ventral striatum) were hypermetabolic compared to values in healthy controls before treatment, but they normalized after L-T4 treatment. This pattern of brain activity before treatment is generally consistent with previous PET findings of specific neuronal correlates of comorbid anxiety symptoms in patients with bipolar depression (Osuch et al., 2000). It is compelling that anxiety of patients with bipolar depression is linked to limbic areas (e.g., amygdala, hippocampus, and parahippocampal gyrus) that are regions of aberrant glucose metabolism in individuals with panic disorder (Bisaga et al., 1998; Miller et al., 2005; Sakai et al., 2005) or other anxiety disorders (e.g., PTSD and social phobia (Tillfors et al., 2001; Charney, 2003). Furthermore trait anxiety symptoms at baseline showed a direct relationship with activity in the superior temporal gyri and dorsal anterior cingulate bilaterally. Increased neural activity in the superior temporal region and insula was found by Chua et al. (1999) to be associated with anticipatory anxiety in healthy subjects. Similarly, Osuch et al. (2000) reported an inverse relationship between anxiety and glucose metabolism in the left superior temporal region and insula. In summary, there are similarities (amygdala, hippocampus, ventral striatum, and anterior cingulate) but also distinctions between metabolic patterns associated with anxiety vs. depression in patients with bipolar disorder. Most notably, the superior temporal gyri, the insula and the parahippocampal gyri have been linked to anxiety, while these areas have not been associated in circuitries in depression. Overall, there were only minor differences in the association of regional brain glucose metabolism with trait as well as with state anxiety measures. As previously reported by others (Ramanaiah et al., 1983), we found a high intercorrelation between the state and trait anxiety scale of the STAI. Therefore, a similar pattern of brain glucose metabolism associated with both trait and state anxiety can be expected, and it is difficult to determine the extent to which residual unshared variance truly maps onto the conceptual state vs. trait dichotomy. The clinical syndrome of excessive thyroid hormone production (thyrotoxicosis) is characterized by excessively high levels of peripheral thyroid hormone and significant changes in behavior. Anxiety is one of the core psychiatric symptoms associated with this condition (Whybrow and Bauer, 2005). In the current study, patients with bipolar depression exhibited improvement of anxiety symptoms during treatment with supraphysiological doses of L-T4. In addition

75

to demonstrating normal peripheral thyroid hormone metabolism, most patients with bipolar disorder who are treated with supraphysiological doses of L-T4, respond differently to the hormone than either those with pre-existing thyroid disease or healthy individuals. The patients tolerate the high doses of L-T4 surprisingly well, demonstrating none of the serious side effects associated with hyperthyroxinaemia (such as loss of bone mineral density and increase of heart rate), even when treated long-term for several years (Stancer and Persad, 1982; Rudas et al., 1999; Bauer et al., 2002, 2004; reviewed in: Bauer et al., 2008). The low rate of side effects in patients with bipolar disorders who are receiving supraphysiological doses of L-T4 contrasts with the high rate typically seen in patients with primary thyroid disease. The reduction in trait anxiety scores during treatment with thyroid hormone was associated with decreases in relative activity in the right amygdala and hippocampus. The results imply that treatment with L-T4 affects neural activity in brain areas previously associated with primary anxiety disorders (Miller et al., 2005). The left middle frontal gyrus and the right dorsal anterior cingulate showed an inverse relationship between changes in state anxiety and changes of relative activity after treatment indicating that a decrease in anxiety was associated with an increase in metabolism of these cortical areas. Vogt (2005) has pointed out to cytoarchitectonic and functional subdivisions within the dorsal anterior cingulate. Limitations of this study include the lack of a control group, the relatively small study group, and the fact that only women were included. It is also uncertain to what degree changes in brain activity and anxiety were a direct response to L-T4 as opposed to a secondary effect of improved mood. Because all of the patients were partial or full responders (Bauer et al., 2005), it is difficult to dissociate direct effects of L-T4 on brain glucose metabolism from secondary effects of improvement in mood. Acknowledgements Supported by Deutsche Forschungsgemeinschaft Grant Ba 1504/3-1 (Dr. Bauer), National Alliance for Research on Schizophrenia and Depression (NARSAD) Young Investigator Award (Dr. Bauer) and the UCLA General Clinical Research Center (GCRC) grant M01-RR00865. References Bauer, M., Hellweg, R., Gräf, K.J., Baumgartner, A., 1998. Treatment of refractory depression with high-dose thyroxine. Neuropsychopharmacology 18, 444–455. Bauer, M., Berghöfer, A., Bschor, T., Baumgartner, A., Kiesslinger, U., Hellweg, R., Adli, M., Baethge, C., Muller-Oerlinhausen, B., 2002. Supraphysiological doses of L-thyroxine in the maintenance treatment of prophylaxis-resistant affective disorders. Neuropsychopharmacology 27, 620–628. Bauer, M., Fairbanks, L., Berghöfer, A., Hierholzer, J., Bschor, T., Baethge, C., Rasgon, N., Sasse, J., Whybrow, P.C., 2004. Bone mineral density during maintenance treatment with supraphysiological doses of levothyroxine in affective disorders: a longitudinal study. Journal of Affective Disorders 83, 183–190. Bauer, M., London, E.D., Rasgon, N., Berman, S.M., Frye, M.A., Altshuler, L., Mandelkern, M.A., Bramen, J., Voytek, B., Woods, R., Mazziotta, J.C., Whybrow, P.C., 2005. Supraphysiological doses of levothyroxine alter regional cerebral metabolism and improve mood in women with bipolar depression. Molecular Psychiatry 10, 456–469. Bauer, M., Goetz, T., Glenn, T., Whybrow, P.C., 2008. The thyroid-brain interaction in thyroid disorders and mood disorders. Journal of Neuroendocrinology 20, 1101–1114. Baxter Jr., L.R., Schwartz, J.M., Phelps, M.E., Mazziotta, J.C., Guze, B.H., Seli, C.E., Gerner, R.H., Sumida, R.M., 1989. Reduction of prefrontal cortex glucose metabolism common to three types of depression. Archives of General Psychiatry 46, 243–250. Bisaga, A., Katz, J.L., Antonini, A., Wrigh, C.E., Margouleff, C., Gorman, J.M., Eidelberg, D., 1998. Cerebral glucose metabolism in women with panic disorder. American Journal of Psychiatry 155, 1178–1183. Charney, D.S., 2003. Neuroanatomical circuits modulating fear and anxiety behaviors. Acta Psychiatrica Scandinavica. Supplementum 417, 38–50. Chua, P., Krams, M., Toni, I., Passingham, R., Dolan, R., 1999. A functional anatomy of anticipatory anxiety. Neuroimage 9, 563–571. Drevets, W.C., 2000. Neuroimaging studies of mood disorders. Biological Psychiatry 48, 813–829. Drevets, W.C., Price, J.L., Simpson Jr., J.R., Todd, R.D., Reich, T., Vannier, M., Raichle, M.E., 1997. Subgenual prefrontal cortex abnormalities in mood disorders. Nature 386, 824–827.

76

M. Bauer et al. / Psychiatry Research: Neuroimaging 181 (2010) 71–76

Duvernoy, H.M., 1999. The human brain, Surface, Three-Dimensional Sectional Anatomy with MRI, and Blood Supply. 2nd ed. Springer, Vienna. Endler, N.S., Kocovski, N.L., 2001. State and trait anxiety revisited. Journal of Anxiety Disorders 15, 231–245. Frank, E., Cyranowski, J.M., Rucci, P., Shear, M.K., Fagiolini, A., Thase, M.E., Cassano, G.B., Grochocinski, V.J., Kostelnik, B., Kupfer, D.J., 2002. Clinical significance of lifetime panic spectrum symptoms in the treatment of patients with bipolar I disorder. Archives of General Psychiatry 59, 905–911. Freeman, M.P., Freeman, S.A., McElroy, S.L., 2002. The comorbidity of bipolar and anxiety disorders: prevalence, psychobiology, and treatment issues. Journal of Affective Disorders 68, 1–23. Ketter, T.A., Kimbrell, T.A., George, M.S., Dunn, R.T., Speer, A.M., Benson, B.E., Willis, M.W., Danielson, A., Frye, M.A., Herscovitch, P., Post, R.M., 2001. Effects of mood and subtype on cerebral glucose metabolism in treatment-resistant bipolar disorder. Biological Psychiatry 49, 97–109. Miller, L.A., Taber, K.H., Gabbard, G.O., Hurley, R.A., 2005. Neural underpinnings of fear and its modulation: implications for anxiety disorders. Journal of Neuropsychiatry and Clinical Neuroscience 17, 1–6. Osuch, E.A., Ketter, T.A., Kimbrell, T.A., George, M.S., Benson, B.E., Willis, M.W., Herscovitch, P., Post, R.M., 2000. Regional cerebral metabolism associated with anxiety symptoms in affective disorder patients. Biological Psychiatry 48, 1020–1023. Ramanaiah, N.V., Franzen, M., Schill, T., 1983. A psychometric study of the State-Trait Anxiety Inventory. Journal of Personality Assessment 47, 531–535. Rauch, S.L., Savage, C.R., Alpert, N.M., Fischman, A.J., Jenike, M.A., 1997. The functional neuroanatomy of anxiety: a study of three disorders using positron emission tomography and symptom provocation. Biological Psychiatry 42, 446–452. Rauch, S.L., Shin, L.M., Wright, C.I., 2003. Neuroimaging studies of amygdala function in anxiety disorders. Annals New York Acad Science 985, 389–410. Rudas, S., Schmitz, M., Pichler, P., Baumgartner, A., 1999. Treatment of refractory chronic depression and dysthymia with high-dose thyroxine. Biological Psychiatry 2, 229–233.

Sakai, Y., Kumano, H., Nishikawa, M., Sakano, Y., Kaiya, H., Imabayashi, E., Ohnishi, T., Matsuda, H., Yasuda, A., Sato, A., Diksic, M., Kuboki, T., 2005. Cerebral glucose metabolism associated with a fear network in panic disorder. NeuroReport 16, 927–931. Simon, N.M., Otto, M.W., Wisniewski, S.R., Fossey, M., Sagduyu, K., Frank, E., Sachs, G.S., Nierenberg, A.A., Thase, M.E., Pollack, M.H., 2004. Anxiety disorder comorbidity in bipolar disorder patients: data from the first 500 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). American Journal of Psychiatry 161, 2222–2229. Stancer, H.C., Persad, E., 1982. Treatment of intractable rapid-cycling manic-depressive disorder with levothyroxine. Archives of General Psychiatry 39, 311–312. Strakowski, S.M., Delbello, M.P., Adler, C.M., 2005. The functional neuroanatomy of bipolar disorder: a review of neuroimaging findings. Molecular Psychiatry 10, 105–116. Talairach, J., Tourneoux, P., 1988. Co-Planar Stereotaxic Atlas of the Human Brain. Thieme, New York. Tillfors, M., Furmark, T., Marteinsdottir, I., Fischer, H., Pissiota, A., Langstrom, B., Fredrikson, M., 2001. Cerebral blood flow in subjects with social phobia during stressful speaking tasks: a PET study. American Journal of Psychiatry 158, 1220–1226. Vogt, B.A., 2005. Pain and emotion interactions in subregions of the cingulate gyrus. Nature Reviews Neuroscience 6, 533–544. Wellcome Department of Cognitive Neurology, London, UK, 2000. Statistical Parametric Mapping. Available at: http://www.fil.ion.ucl.ac.uk/spm/. Accessed 25th January. Whybrow, P.C., Bauer, M., 2005. Behavioral and psychiatric aspects of thyrotoxicosis. In: Braverman, L.E., Utiger, R.D. (Eds.), The Thyroid. A Fundamental and Clinical Text, 9th ed. Lippincott-Raven, Philadelphia, pp. 644–650. Wittchen, H.U., Zhao, S., Kessler, R.C., Eaton, W.W., 1994. DSM-III-R generalized anxiety disorder in the National Comorbidity Survey. Archives of General Psychiatry 51, 355–364. Woods, R.P., Mazziotta, J.C., Cherry, S.R., 1993. MRI-PET registration with automated algorithm. Journal of Computer Assisted Tomography 17, 536–546.