Metab Brain Dis (2006) 21:241–252 DOI 10.1007/s11011-006-9009-2 ORIGINAL PAPER A comparison of the effects of citalopram and moclobemide on resting brain perfusion in social anxiety disorder J. M. Warwick · P. Carey · G. Van der Linden · C. Prinsloo · D. Niehaus · S. Seedat · P. Dupont · D. J. Stein Received: 30 June 2005 / Accepted: 10 October 2005 / Published online: 21 September 2006
C Springer Science+Business Media, Inc. 2006 Abstract Introduction: The serotonin specific reuptake inhibitor (SSRI) citalopram and the reversible mono-amine oxidase-A inhibitor (RIMA) moclobemide have both been used successfully for the treatment of social anxiety disorder (SAD). In this study we investigate the effects of these compounds on resting brain function using single photon emission computed tomography (SPECT). Methods: Subjects meeting DSM-IV criteria for SAD underwent regional cerebral blood flow (rCBF) SPECT using Tc-HMPAO at baseline and after 8 weeks of treatment with either citalopram or moclobemide. Using statistical parametric mapping brain SPECT studies were analysed to determine the effects of treatment on rCBF, to compare the effects of citalopram and moclobemide, and to detect correlations between changes in rCBF and clinical response. Results: Subjects received citalopram (n = 17) or moclobemide (n = 14) as therapy. Subjects in both treatment groups demonstrated a significant improvement of SAD symptoms as measured by the Liebowitz Social Anxiety Scale total score. All subjects demonstrated a decrease in rCBF in the insulae post therapy. Subjects receiving citalopram had decreased superior cingulate rCBF after therapy compared to those receiving moclobemide. Conclusion: Both SSRI’s and RIMA’s decreased rCBF in the insulae during treatment of SAD; an effect that may be consistent with the role of these regions in processing internal somatic cues evoked by emotional stimuli. Citalopram had a greater effect on J. M. Warwick · P. Carey · G. Van der Linden · C. Prinsloo · D. Niehaus · S. Seedat · D. J. Stein MRC Unit for Stress and Anxiety Disorders, Faculty of Health Sciences, Stellenbosch University, Cape Town, South Africa P. Dupont Nuclear Medicine, Katolieke Universiteit, Leuven, Belgium J. M. Warwick (
) Nuclear Medicine, Tygerberg Hospital, Francie van Zijl Drive, Tygerberg, 7505 Cape Town, South Africa e-mail: jw@sun.ac.za Springer 242 Metab Brain Dis (2006) 21:241–252 superior cingulate perfusion, an effect that is consistent with evidence of high levels of 5-HT transporters in this region. Keywords Social phobia . SPECT . Citalopram . Moclobemide . Insula . Cingulate Introduction Social anxiety disorder (SAD) is recognized as a discrete anxiety disorder in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association, 1994). It is a condition characterised by fears of social interaction and performance situations. These fears cause subjects to avoid these situations or endure them with extreme distress. Unlike normal social anxiety, SAD results in significant distress and impairment (Wittchen and Beloch, 1996). SAD, the most prevalent of the major anxiety disorders with a lifetime prevalence of 10–15%, is associated with significant comorbidity and morbidity (Ballenger, 1998; Magee et al., 1996, Weiller et al., 1996). The underlying neurocircuitry that mediates SAD is poorly understood, but may involve regions implicated in anxiety (LeDoux, 1998; Nutt et al., 1998). Thus, neuroimaging studies of SAD demonstrated increased activity in the amygdalo-hippocampal complex during anticipation (Lorderbaum et al., 2004; Tillfors et al., 2002), public speaking (Tillfors et al., 2001), and conditioned response (Veit et al., 2002; Birbaumer et al., 1998; Schneider et al., 1999) paradigms. The amygdalae are believed to play a role in fear conditioning whereas the hippocampi may be responsible for processing the context of fear conditioning. Simultaneously activity in many cortical areas appears to decrease during anxiety provocation, perhaps reflecting impaired cognitive processing at these times (Lorderbaum et al., 2004; Tillfors et al., 2001; Tillfors et al., 2002; Van Ameringen et al., 2004). A growing body of literature has also implicated specific neurotransmitter systems in SAD. In particular, evidence has pointed to a role of the dopaminergic and serotonergic systems (Stein et al., 2002). Selective serotonin reuptake inhibitors (SSRI’s) are effective for the treatment of SAD (Van der Linden et al., 2000a), and as a class, represent the first-line pharmacotherapy in SAD. SSRIs are thought to exert their effect in SAD by decreasing the uptake of serotonin in the synaptic cleft, thus enhancing serotonergic transmission. Serotonergic projections appear to have an important influence on the pathways that mediate anxiety (Coplan and Lydiard, 1998), although the exact effects of SSRIs on this neurocircuitry remain unclear. Similarly, there is evidence demonstrating the efficacy of monoamine oxidase inhibitors, including the reversible inhibitors of monoamine oxidase (RIMAs) in the treatment of SAD (Liebowitz et al., 1992). These drugs have a less specific effect on neurotransmission; by inhibiting the enzyme monoamine oxidase, they result in an increase in intrasynaptic levels of a range of monoamines (noradrenaline, dopamine, and serotonin). Meta-analyses have suggested that SSRIs may be more effective than RIMAs in the treatment of SAD (Van der Linden et al., 2000a), and the latter agents are therefore more often used in clinical situations where tolerability is a particular asset. While both SSRI’s and RIMA’s have demonstrated clinical efficacy, little is known about the neuro-anatomical mechanisms underlying the efficacy of these drugs in SAD. In a previous region of interest (ROI) based study, after 8 weeks of citalopram treatment, SAD patients had a decrease in resting rCBF in the temporal, frontal, and cingulate cortex on the left side (Van der Linden et al., 2000b). Furmark et al. described attenuation of the rCBF response to a public speaking task bilaterally in the amygdala, hippocampus, Springer Metab Brain Dis (2006) 21:241–252 243 and the periamygdaloid, rhinal, and parahippocampal cortices after 9 weeks of treatment with either SSRI or cognitive behavioural therapy (Furmark et al., 2002). The authors are not aware of any published work describing the effects of MAO inhibitors or RIMA’s on rCBF. A comparison of the effects on rCBF of SAD treatment with a SSRI and with a RIMA could provide insights into their mechanisms of action in this condition. In this study the effects on resting rCBF of the highly selective SSRI citalopram and the RIMA moclobemide were compared and contrasted using brain SPECT. Data from 12 of these patients, receiving only citalopram, has previously been reported using a ROI analysis (Van der Linden et al., 2000b); in this study we extended the sample receiving citalopram, added subjects receiving moclobemide, and used a voxel based analysis. Methods Subjects Adult subjects with a primary diagnosis of generalised SAD were recruited from the Anxiety Disorders Clinic of our tertiary hospital. All subjects were interviewed with the Structured Clinical Interview for the Diagnosis of Axis-I Disorders to ascertain diagnosis according to DSM-IV criteria (DSM IV) (First et al., 1996). Subjects with other primary psychiatric disorders, significant medical illness or a neurological condition were excluded. A few patients had comorbid psychiatric disorders (2–panic disorder, 1–trichotillomania (TTM), 1–generalised anxiety disorder (GAD) and body dysmorphic disorder, and 1–GAD and TTM). These conditions were considered secondary in terms of temporal course, symptom severity, and associated distress and the subjects were therefore included. All data was acquired after receiving approval from the Institutional Review Board of our University and all patients gave informed written consent after a full explanation of the possible risks and benefits. Pharmacotherapy All patients underwent treatment for 8 weeks with either the selective serotonin reuptake inhibitor citalopram, or the reversible inhibitor of monoamine oxidase-A moclobemide. Subjects receiving citalopram were initiated on a dosage of 20 mg daily for the first two weeks and then increased to 40 mg daily for the remaining 6 weeks. Moclobemide therapy was initiated at 600 mg/day for first two weeks and then increased to 900 mg/day for the remaining six weeks. Measures Ratings were carried out at baseline and every two weeks until the completion of pharmacotherapy. Social anxiety symptoms, depression, and disability were rated using the Liebowitz Social Anxiety Scale (LSAS), Montgomery-Asberg Depression Rating Scale (MADRS), and Sheehan Disability Scale (SDS) respectively. (Liebowitz, 1987; Montgomery and Asberg, 1979; Sheehan, 1983). Springer 244 Metab Brain Dis (2006) 21:241–252 SPECT imaging Single photon emission computed tomography (SPECT) was conducted before and after pharmacotherapy. Subjects lay supine in a quiet dimly lit room for 30 min prior to injection of the radiopharmaceutical. Apart from administration of the injection by a physician, they remained alone in the room during this period. Subjects were asked to remain at rest for 10 min after the injection of the radiopharmaceutical to allow uptake of the radiopharmaceutical in the brain. An injection of 555 MBq (15 mCi) of technetium-99 m hexamethylpropylene amine oxime (Tc-99 m HMPAO) was given into an arm vein through a previously placed intravenous cannula. After completion of the 10 min rest period described above, SPECT imaging of the brain was performed, with the subject’s head supported by a headrest, using a dual detector gamma camera (Elscint Helix, GE Medical Systems, USA) equipped with fanbeam collimators. Data were acquired in the step-and-shoot mode, using a 360 degree circular orbit, with the detectors of the gamma camera as close as possible to the subject’s head. The radius of rotation was noted for each subject and the same measurements were used for the follow-up study. Data were acquired using a 128 × 128 image matrix in 3 degree steps of 15 seconds per step. Data were reconstructed by filtered backprojection, using a Metz filter (power = 5, FWHM = 14 mm). The Chang method (µ = 0.11/ cm) was used for attenuation correction (Chang, 1978). The final reconstructed voxel size was 1.7 × 1.7 × 3.9 mm3 . Image files were converted from interfile to analyze format using conversion software (Medcon, Erik Nolf, UZ Ghent, Belgium). Spatial preprocessing Statistical analyses were conducted on a voxel-by-voxel basis using Statistical Parametric Mapping (SPM99, Wellcome Department of Cognitive Neurology, UK) (Friston et al., 1991). The realign function was used to co-register baseline and post-treatment SPECT images for each subject and to generate a mean image for each subject. Realigned images were then normalised to the Montreal Neurological Institute (MNI) standard anatomical space with 4 × 4 × 4 mm3 voxels, and to a value of 50 using proportional scaling. This was achieved using the transform function from the realigned mean image for each subject, to the template image in MNI space using 12 affine transformations and 7 × 8 × 7 non-linear basis functions. The normalised images were then smoothed using a 3D Gaussian kernel with a FWHM of 12 mm. Analysis A multi-group study design was performed using 2 groups (citalopram therapy and moclobemide therapy) with 2 conditions each (pre- and post-treatment). Contrasts were applied to detect: (i) areas of significant change post-treatment compared to pre-treatment, and (ii) areas of relative change in subjects receiving citalopram compared to those receiving moclobemide. A second study design, using the change in LSAS as a covariate, was performed to detect correlations between the change in LSAS and the change in rCBF as a result of therapy. Contrasts were applied to detect positive and negative correlations. Springer Metab Brain Dis (2006) 21:241–252 Table 1 245 Mean and standard deviation of various scales Baseline Post therapy % change p∗ Total Liebowitz Social Anxiety Scale (LSAS) scores at baseline and following pharmacotherapy Citalopram (n = 17) 95.2 ± 21.4 76.1 ± 26.9 20.1 <0.01 Moclobemide (n = 14) 107.6 ± 20.7 69.4 ± 20.1 35.5 <0.001 Montgomery-Asberg Depression Rating Scale (MADRS) scores at baseline and following pharmacotherapy Citalopram (n = 17) 14.1 ± 5.3 7.1 ± 4.4 49.6 <0.0001 Moclobemide (n = 14) 7.7 ± 6.5 3.8 ± 2.6 50.6 <0.01 Total Sheehan Disability Scale (SDS) scores at baseline and following pharmacotherapy Citalopram (n = 17) 17.9 ± 4.7 12.6 ± 7.3 29.6 <0.001 Moclobemide (n = 14) 18.9 ± 4.4 13.6 ± 6.0 28.0 <0.01 ∗ One-tailed paired Student’s t-test. An uncorrected p-value of p < 0.001 was chosen as a threshold for statistical significance. A spatial extent threshold of 10 voxels was also used at all times. Clusters were located to anatomical regions using MRIcro software (Chris Rorden, Nottingham University, UK). Results Thirty-one subjects (21 male, 10 female) with a mean (SD) age of 33 (9) years completed the study. Seventeen subjects received citalopram and 14 subjects received moclobemide. The two groups were clinically similar, although those receiving citalopram had higher MADRS scores at baseline (p < 0.01). All but two subjects, one on alprazolam 1 mg/d and another on propranolol 20 mg/d, were free of any other psychotropic medication during the study. Clinical changes with pharmacotherapy for each drug are provided in Table 1. Pharmacotherapy using both drugs was associated with a significant reduction in social anxiety symptom severity, depression symptoms, and disability. The results of the SPM analysis are summarised in Table 2. Comparison of pre- and post-treatment scans for the whole group showed decreased activity in 2 significant clusters located in the regions of right and left insulae. (Fig. 1) Comparing the effects of citalopram with moclobemide demonstrates a significantly greater decrease in perfusion in the superior cingulate in those subjects receiving citalopram. (Fig. 2). Using the change in LSAS as a covariate, a significant correlation was detected between the LSAS change score and decreased rCBF in the posterior cingulate. On closer examination of this region it was found that across patients when comparing the pre- and post-therapy scans, the response to therapy in this region was heterogenous with a mixture of decreasing and increasing rCBF. This correlation therefore represented a mixture of a progressively smaller increases in rCBF (in subjects with increased rCBF with treatment) and a progressively larger decrease in rCBF (in subjects with decreased rCBF with treatment), with increasing LSAS change score. Conversely two other clusters, in the right mid-temporal cortex and the cuneus, showed a significant correlation between change in LSAS and increased rCBF following therapy. (Fig. 3). Furthermore, although not attaining the high level of significance required for the voxelwise analyses above, an analysis of the individual voxels with the most significant deactivations in the insulae of the combined group following therapy, demonstrated a positive correlation with the change in LSAS (p < 0.05). Springer 246 Metab Brain Dis (2006) 21:241–252 Table 2 Changes in rCBF following pharmacotherapy, differences between citalopram and moclobemide therapy, and correlations between change in LSAS and rCBF changes with therapy Region Decreased for both drugs: Right insula Left insula Decreased in citalopram, not in moclobemide Mid cingulate Correlation between change in LSAS and decrease in rCBF Posterior cingulate Right insula∗ Left insula∗ Correlation between change in LSAS and increase in rCBF Right mid-temporal cortex Cuneus MNI coordinates x,y,z (mm) Voxels t 36, 12, − 4 − 44, 4, 4 27 11 5.07 4.21 4, 12, 44 18 4.69 − 8, − 36, 24 36, 12, − 4 − 44, 4, 4 16 4.39 2.05 2.22 60, − 52, 16 − 4, − 84, 24 10 21 4.40 3.90 ∗ Not significant for voxelwise analysis of whole brain volume, but p < 0.05 for single voxels at sites of maximum insular deactivation with therapy. Discussion A first finding in this study was that resting rCBF in subjects with social anxiety disorder decreased in the insulae bilaterally following therapy with both the SSRI citalopram and the RIMA moclobemide. This was also noted when looking at the drugs separately (results not shown), although this did not reach statistical significance, probably due to the small size of the individual treatment groups. Furthermore, when looking at the voxels with the most significant deactivation in the left and right insulae, there was a significant relationship between the magnitude of the deactivation and the clinical response based on a change in the LSAS. Thus, insular deactivation may not merely reflect treatment with the medications here, but may be directly related to effective pharmacotherapy. It is interesting that despite differences in their pharmacological actions both of these drugs were both clinically effective and both had common effects on rCBF. Little is known about the effects of SAD on resting rCBF. Only one study has been published comparing resting rCBF in SAD patients with a control group (Stein and Leslie, 1996). This study found no differences between the two groups, but only limited areas of the brain were examined using manually placed regions of interest. In particular the study did not focus on insular perfusion. It is possible that resting rCBF may be increased in the insulae in SAD patients, with normalisation following successful pharmacotherapy. However there is no experimental data to substantiate or refute this. Other studies have demonstrated that SSRI’s alter brain perfusion (Carey et al., 2004; Furmark et al., 2002; Geday et al., 2005) and metabolism (Smith et al., 2002; New et al., 2004). The overlapping effects on rCBF of these two drugs with differing primary mechanisms of action, suggests that at least some of their direct or secondary actions seem to overlap and to effect a clinical response in SAD. Brain perfusion is however an indirect measure of neuronal metabolism and neuronal activity. Although these processes can become decoupled under certain circumstances, they are normally linked (Kessler, 2003). Thus, it Springer Metab Brain Dis (2006) 21:241–252 247 Fig. 1 Two clusters of significantly decreased rCBF in the left and right insulae in all subjects following pharmacotherapy with citalopram or moclobemide. The clusters are superimposed onto a representative T1-weighted MRI study in MNI space. (Courtesy of MRIcro, Chris Rorden, Nottingham University, UK) can be speculated that in SAD the effects of SSRI’s and RIMA’s on neuronal function may converge to a common pathway involving the insulae. Previous work in SAD has implicated the insulae with demonstration of increased activity during anticipation (Lorderbaum et al., 2004) and conditioned response (Veit et al., 2002) paradigms. In apparent contrast, decreased activity during a public speaking task has been demonstrated (Tillfors et al., 2001). It is possible that the insulae activate the amygdalohippocampal complexes during anticipation of an unpleasant task or while a conditioned response is evoked (Veit et al., 2002). However during exposure to the feared situation, the insulae no longer play this role. After successful pharmacotherapy, social stimuli are less likely to be experienced as aversive, and there is apparently a decrease in baseline activity in this region. We were unable to demonstrate an effect of SSRI treatment on the amygdalohippocampal complex in SAD, perhaps because of the relatively poor spatial resolution of SPECT compared with PET. A primary effect of antidepressants on the insulae cannot Springer 248 Metab Brain Dis (2006) 21:241–252 Fig. 2 Single cluster of significantly decreased rCBF in the superior cingulum in subjects receiving citalopram compared to those receiving moclobemide. The cluster is superimposed onto a representative T1-weighted MRI study in MNI space. (Courtesy of MRIcro, Chris Rorden, Nottingham University, UK) however be ruled out on the basis of the data at hand. Changes in the insulae after SSRIs may also not be limited to SAD; decreased resting glucose metabolism was noted in the right insula of elderly patients with depression following 8 weeks of citalopram therapy (Smith et al., 2002). This raises the question of whether specific serotonin receptors in the insulae are involved in the response to SSRIs. At the same time, caution is required in suggesting that the citalopram or moclobemide act directly on specific receptors in particular regions in order to change perfusion there. These agents have a range of secondary effects in addition to their initial effects on synaptic neurotransmitters. Despite this, autoradiographic work has demonstrated high densities of serotonin transporters in the limbic system and the insulae (Varnäs et al., 2004). Studies with serotonergic ligands in SAD might be useful in shedding light on this issue. Springer Metab Brain Dis (2006) 21:241–252 249 Fig. 3 Single cluster of significant correlation between magnitude of change of LSAS and decrease in rCBF in the region of the posterior cingulate. Two clusters of significant correlation between magnitude of change of LSAS and increase in rCBF in the right mid-temporal cortex and cuneus. The clusters are superimposed onto a representative T1-weighted MRI study in MNI space. (Courtesy of MRIcro, Chris Rorden, Nottingham University, UK) A second finding was that patients receiving citalopram demonstrated a significant decrease in resting perfusion to the mid-cingulate compared to those receiving moclobemide. In the group as a whole, there was a correlation between improved symptoms and deactivation of the posterior cingulate. Deactivation of the cingulate is consistent with an earlier ROI based report of SAD patients treated with citalopram using a sub-set of the subjects used in this study (Van der Linden et al., 2000b) and with a mixed population of anxiety disorders in a population that also included a subset of the patients in the current study (Carey et al., 2004). The greater effect of citalopram on the cingulate is consistent with innervation of cingulate cortex by inhibitory serotonergic neurons arising from the raphe nuclei in the brainstem, with autoradiographic studies demonstrating a high concentration of Springer 250 Metab Brain Dis (2006) 21:241–252 serotonin transporters in this region (Varnäs et al., 2004), and imaging studies showing a high level of occupancy of the serotonin transporter at therapeutic levels of SSRI’s (Kent et al., 2002). It has been suggested that therapeutic interventions in SAD may exert their effect by suppression of neuronal activity in the amygdalo-hippocampal and surrounding cortical regions, including the cingulate (Furmark et al., 2002). The amygdalae receive inhibitory inputs from the cingulate that are believed to mediate the extinction of fear conditioning. Consequently amygdala hyperactivity in SAD may result in a compensatory increase in cingulate activity that is no longer needed following successful pharmacotherapy. A number of limitations of the current study can be noted. First, two subjects, one on alprazolam 1 mg/d and another on propranolol 20 mg/d, were receiving other psychotropic medication during the study. However some evidence exists suggesting that rCBF is unaffected by the chronic administration of these drugs (Roy-Byrne et al., 1993; Madsen et al., 1990). Second, the difference in the severity of the baseline depression symptoms between the 2 groups is a potential confound, so that a more marked decrease in cingulate perfusion in the citalopram group may be a reflection of the treatment of more severe depressive symptoms in this group. Third, the relatively small sample size and the limitations of SPECT may have lead to an inability to detect and identify true differences, as well as false positive findings. The choice of significance threshold is a compromise to attempt to minimise the presence of type 1 and type 2 errors. Nevertheless, the findings contribute to a growing literature on the functional imaging of anxiety disorders. Future work will also need to assess particular neuroreceptor systems before and after pharmacotherapy of SAD. In particular, reasonably available and validated SPECT ligands exist that may be used to evaluate the serotonergic and dopaminergic systems in these patients (de Win et al., 2005). References American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders, 4th edn. American Psychiatric Association, Washington, DC Ballenger JC (1998) Focus on social anxiety disorder. J Clin Psychiatry 59(suppl. 17):3 Birbaumer N, Grodd W, Diedrich O, Klose U, Erb M, Lotze M, Schneider F, Weiss U, Flor H (1998) fMRI reveals amygdala activation to human faces in social phobics. Neuroreport 9:1223–1226 Carey PD, Warwick J, Niehaus DJ, Van der Linden G, Van Heerden BB, Harvey BH, Seedat S, Stein DJ (2004) Single photon emission computed tomography (SPECT) of anxiety disorders before and after treatment with citalopram. BMC Psychiatry 4:30 Chang LT (1978) A method for attenuation correction in radionuclide computed tomography. IEEE Trans Nucl Sci 25:638–643 Coplan JD, Lydiard RB (1998) Brain circuits in panic disorder. Biol Psychiatry 44:1264–1276 Deakin JWF et al (1991) J Psychopharmacol 5:305–315 de Win MM, Habraken JB, Reneman L, van den Brink W, den Heeten GJ, Booij J (2005) Validation of [(123)I]beta-CIT SPECT to assess serotonin transporters in vivo in humans: a double-blind, placebocontrolled, crossover study with the selective serotonin reuptake inhibitor citalopram. Neuropsychopharmacology 30:996–1005 First MB, Spitzer RL, Gibbon M, Williams JBW (1996) Structured Clinical Interview for DSM-IV Axis I Disorders—Clinician Edition. Biometrics Research Department, New York State Research Institute, NY Friston KJ, Frith CD, Liddle PF, Frackowiak RS (1991) Comparing functional (PET) images: the assessment of significant change. J Cereb Blood Flow Metab 11:690–699 Furmark T, Tillfors M, Marteinsdottir I, Fischer H, Pissiota A, Långström B, Fredrikson M (2002) Common changes in cerebral blood flow in patients with social phobia treated with citalopram or cognitivebehavioural therapy. Arch Gen Psychiatry 59:425–433. Springer Metab Brain Dis (2006) 21:241–252 251 Geday J, Hermansen F, Rosenberg R, Smith DF (2005) Serotonin modulation of cerebral blood flow measured with positron emission tomography (PET) in humans. Synapse 55:224–229 Graeff FG, Guimaraes FS, De Andrade TG, Deakin JF (1996) Role of 5-HT in stress, anxiety, and depression. Pharmacol Biochem Behav 54:129–141 Kent JM, Coplan JD, Lombardo I, Hwang D, Huang Y, Mawlawi O, Van Heertum RL, Slifstein M, Abi-Dargham A, Gorman JM, Laruelle M (2002) Occupancy of brain serotonin transporters during treatment with paroxetine in patients with social phobia: a positron emission tomography study with [11C]McN 5652. Psychopharmacology 164:341–348 Kessler RM (2003) Imaging methods for evaluating brain function in man. Neurobiol Aging 24:S21–35 LeDoux J (1998) Fear and the brain: where have we been and where are we going? Biol Psychiatry 44:1229– 1238 Liebowitz MR (1987) Social Phobia. Mod Probl Pharmacopsychiat 22:141–173 Liebowitz MR, Schneier FR, Campeas R, Hollander E, Hatterer J, Fyer A, Gorman J, Papp L, Davies S, Gully R, Klein DF (1992) Phenelzine versus atenolol in social phobia: a placebo-controlled trial. Arch Gen Psychiatry 49:290–300 Lorderbaum JP, Kose S, Johnson MR, Arana GW, Sullivan LK, Hamner MB, Ballenger JC, Lydiard RB, Brodrick PS, Bohning DE, George MS (2004) Neural correlates of speech anticipatory anxiety in generalised social phobia. Neuroreport 15:2701–2705 Madsen PL, Vorstrup S, Schmidt JF, Paulson OB (1990) Effect of acute and prolonged treatment with propranolol on cerebral blood flow and cerebral oxygen metabolism in healthy volunteers. Eur J Clin Pharmacol. 39:295–297 Magee WJ, Eaton WW, Wittchen HU, McGonagle KA, Kessler RC (1996) Agoraphobia, simple phobia and social phobia in the National Comorbidity Survey. Arch Gen Psychiatry 53:159–168 Montgomery SA, Asberg M (1979) A new depression scale designed to be sensitive to change. Br J Psychiatry. 134:382–389 New AS, Buchsbaum MS, Hazlett EA, Goodman M, Koenigsberg HW, Lo J, Iskander L, Newmark R, Brand J, O’Flynn K, Siever LJ (2004) Fluoxetine increases relative metabolic rate in prefrontal cortex in impulsive aggression. Psychopharmacology (Berl) 176:451–458 Nutt DJ, Bell CJ, Malizia AL (1998) Brain mechanisms of social anxiety disorder. J Clin Psychiatry 59(suppl. 17):4–9 Roy-Byrne P, Fleishaker J, Arnett C, Dubach M, Stewart J, Radant A, Veith R, Graham M (1993) Effects of acute and chronic alprazolam treatment on cerebral blood flow, memory, sedation, and plasma catecholamines. Neuropsychopharmacology 8:161–169 Schneider F, Weiss U, Kessler C, Muller-Gartner HW, Posse S, Salloum JB, Grodd W, Himmelmann F, Gaebel W, Birbaumer N (1999) Subcortical correlates of differential classical conditioning of aversive emotional reactions in social phobia. Biol Psychiatry 45:863–871 Sheehan D (1983) The anxiety disease, Scribners, New York Smith GS, Kramer E, Hermann CR, Goldberg S, Ma Y, Dhawan V, Barnes A, Chaley T, Belakhleff A, Laghrissi-Thode F, Greenwald B, Eidelberg D, Pollock BG (2002) Acute and chronic effects of citalopram on cerebral glucose metabolism in geriatric depression. Am J Geriatr Psychiatry 10:715–723 Stein DJ, Westenberg HGM, Liebowitz MR (2002) Social anxiety disorder and generalised anxiety disorder: serotoninergic and dopaminergic neurocircuitry. J Clin Psychiatry 63(suppl 6):12–19 Stein MB, Leslie WD (1996) A brain single photon-emission computed tomography (SPECT) study of generalized social phobia. Biol Psychiatry 39:825–828 Tillfors M, Furmark T, Marteinsdottir I, Fischer H, Pissiota A, Långström B, Fredrikson M (2001) Cerebral blood flow in subjects with social phobia during stressful speaking tasks: a PET study. Am J Psychiatry 158:1220–1226 Tillfors M, Furmark T, Marteinsdottir I, Fredrikson M (2002) Cerebral blood flow during anticipation of public speaking in social phobia: a PET study. Biol Psychiatry 52:1113–1119 Van Ameringen M Mancini C, Szechtman H, Nahmias C, Oakman JM, Hall GBC, Pipe B, Farvolden P (2004) A PET provocation study of generalised social phobia. Neuroimaging 132:13–18 Van der Linden GJ, Stein DJ, Van Balkom AJ (2000) The efficacy of the selective serotonin reuptake inhibitors for social anxiety disorder (social phobia): a meta-analysis of randomised controlled trials. Int Clin Psychopharmacol 15(suppl. 2):S15–S23 Van der Linden G, Van Heerden B, Warwick J, Wessels C, Van Kradenburg J, Zungu-Dirwayi N, Stein DJ (2000) Functional brain imaging and pharmacotherapy in social phobia: single photon emission computed tomography before and after treatment with the selective serotonin reuptake inhibitor citalopram. Prog Neuro-psychopharmacol and Biol Psychiat 24:419–438 Varnäs K, Halldin C, Hall H (2004) Autoradiographic distribution of serotonin transporters and receptor subtypes in human brain. Hum Brain Mapp 22:246–260 Springer 252 Metab Brain Dis (2006) 21:241–252 Veit R, Flor H, Erb M, Hermann C, Lotze M, Grodd W, Birbaumer N (2002) Brain circuits involved in emotional learning in antisocial behaviour and social phobia in humans. Neuroscience Letters 328:233– 236 Wittchen HU, Beloch E (1996) The impact of social phobia on quality of life. Int Clin Psychopharmacol 11(suppl. 3):15–23 Weiller E, Bisserbe J-C, Boyer P, Lepine JP, Lecrubier Y (1996) Social phobia in general health care: An unrecognised undertreated disabling disorder. Br J Psychiatry 168:169–174 Springer