Journal of Affective Disorders 169 (2014) 105–111

Contents lists available at ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Research report

Thalamocortical abnormalities in auditory brainstem response patterns distinguish DSM-IV bipolar disorder type I from schizophrenia Mia Sköld a,1, Johan Källstrand a,1, Sara Nehlstedt a, Annelie Nordin b, Sören Nielzén a, Jens Holmberg a, Rolf Adolfsson b,n a b

SensoDetect AB, Lund, Sweden Division of Psychiatry, Department of Clinical Sciences, Umeå University, Umeå, Sweden

art ic l e i nf o

a b s t r a c t

Article history: Received 21 February 2014 Received in revised form 14 July 2014 Accepted 3 August 2014 Available online 13 August 2014

Background: Bipolar disorder type I (BP-I) belongs to a spectrum of affective disorders that are expressed in many different ways and therefore can be difficult to distinguish from other conditions, especially unipolar depression, schizoaffective disorder, schizophrenia (SZ), but also anxiety and personality disorders. Since early diagnosis and treatment have shown to improve the long-term prognosis, complementary specific biomarkers are of great value. The auditory brainstem response (ABR) has previously been applied successfully to identify specific abnormal ABR patterns in SZ and Asperger syndrome. Methods: The current study investigated the early auditory processing of complex sound stimuli e.g. forward masking, in BP-I compared to SZ patients. The ABR curves of BP-I patients (n ¼ 23) and SZ patients (n ¼20) were analyzed in terms of peak amplitudes and correlation with an ABR norm curve based on a non-psychiatric control group (n ¼ 20). Results: BP-I patients had significantly higher wave III (p¼ 0.0062) and wave VII (p ¼0.0472) amplitudes compared with SZ patients. Furthermore, BP-I patients, and to a lesser extent SZ patients, showed low correlation with the norm ABR curve in the part of the curve comprising waves VI–VII. Limitations: Sample size was relatively small and study groups were not matched for age and gender. Conclusions: BP-I patients showed specific aberrances, specifically in the latter part of the ABR curve, implicating abnormalities in thalamocortical circuitry. The abnormal ABR wave patterns significantly separated BP-I patients from SZ patients suggesting that ABR might serve as a biomarker for BP-I. & 2014 Elsevier B.V. All rights reserved.

Keywords: Auditory brainstem response Electrophysiology Thalamus Biomarker Bipolar disorder Schizophrenia

1. Introduction Bipolar disorder type I (BP-I) is a serious psychiatric disorder characterized by alternating cycles of mania and depression. Symptoms usually appear in adolescence or early adulthood but the diagnosis is often not recognized until 5–10 years later (Baldessarini et al., 2007; Hauser et al., 2007). The importance of early diagnosis and intervention is stressed by the fact that delayed treatment may be less effective (Swann et al., 1999). The most deleterious effects of treatment delay are seen in children and adolescents who will spend more time in depressive states, show greater severity of depression and more number of episodes, all of which are risk factors for poor outcome in adulthood (Post et al., 2010).

n

Corresponding author. Tel.: þ 46 705907320. E-mail address: [email protected] (R. Adolfsson). 1 Contributed equally to this work.

http://dx.doi.org/10.1016/j.jad.2014.08.002 0165-0327/& 2014 Elsevier B.V. All rights reserved.

BP-I mimics many other mental disorders exhibiting instability in mood and psychomotor functioning, appearing periodically or as in a chronic fluctuating manner over the course of the disease. The most common misdiagnoses are those with pronounced affective and/or psychotic symptomatology, e.g. unipolar depression, schizophrenia (SZ) and schizoaffective disorder (Bowden, 2001; Murray et al., 2004; for review of diagnostic validity, see Kendell and Jablensky, 2003). Concerning bipolar disorders, voices are raised against the huge increase of the diagnosed bipolar disorders over the past decades, emphasizing the need for objective measures of bipolar disorders, not only for BP-I, but also the spectrum outside the classic BP-I type (Mitchell, 2012). The pathophysiology underlying bipolar symptoms is still poorly understood. Magnetic resonance imaging (MRI) studies have indicated abnormalities in the prefrontal cortex as well as limbic/subcortical structures involved in emotional regulation including amygdala and hippocampus, although the results are inconsistent (reviewed in Strakowski et al., 2005). Thalamus, that holds a key role in emotional regulation, has in most studies not shown any significant volume changes in BP-I patients (reviewed

106

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

in Ng et al., 2009), although Strakowski et al. showed an increased thalamus volume (Strakowski et al., 1999) and one study showed a decreased thalamus volume in BP-I patients not treated with lithium (Radenbach et al., 2010). Neurophysiological techniques e.g. auditory event-related potentials (ERPs) have been extensively investigated in SZ using the electroencephalogram (EEG) technique, but have been far less studied in bipolar disorder. For example, P50 suppression that measures sensory gating i.e. the individual's ability to selectively attend to specific auditory stimuli and ignore background noise, is deficient in schizophrenia (Bramon et al., 2004) and diminished P50 suppression has also been reported in bipolar disorder patients with a history of psychosis (Olincy et al., 2005; Schulze et al., 2007) and unaffected first-degree relatives (Schulze et al., 2007), indicating a reduced capacity to ignore background noise. Mismatch negativity (MMN) that reflects a pre-attentive stage of auditory processing, however, is suggested to be normal in bipolar disorder in contrast to schizophrenia (Catts et al., 1995; Hall et al., 2009; Umbricht et al., 2003). Thus, to a large extent similar deficiencies have been observed in SZ and bipolar disorder using electrophysiological methods, although differences are also reported. The auditory brainstem response technique is frequently used to investigate the lower part of the auditory pathway. The technique is, in contrast to EEG and several other imaging techniques, unaffected by level of consciousness and attention of the patient (Chiappa, 1997). The ABR consists of the waves of the auditory evoked potentials that occur earlier than 10 ms poststimulus. These waves are designated waves I–VII. Waves I and II are generated by the ipsilateral and central portion of the auditory nerve, respectively. Wave III is generated by the cochlear nucleus, wave IV is believed to originate from superior olivary complex (SOC) and wave V to represent activity at the levels of lateral lemniscus and inferior colliculus. The contralateral IC may also contribute to waves VI and VII, although thalamic origin (e.g. medial geniculate body) is also suggested (Burkard et al., 2007). ABR studies investigating schizophrenia have shown contradicting results. Several studies have found significant differences between SZ patients and healthy controls when investigating parameters such as latencies, amplitudes, missing peaks and differences between left and right ear (Grillon et al., 1990; Hayashida et al., 1986; Igata et al., 1994; Lindström et al., 1987, 1990), whereas other studies found no significant differences (Brecher and Begleiter, 1985; Josiassen et al., 1980; McKay et al., 2000; Pfefferbaum et al., 1980). Most of the studies reporting differences did so for a subgroup of patients, which may be due to heterogeneity among SZ patients. It has also been suggested that brainstem pathology may be subtle and difficult to identify with routine ABR (Grillon et al., 1990). Using forward masking stimuli and a more elaborate analysis, specific abnormal ABR patterns have been indicated in SZ (Källstrand et al., 2012) and Asperger syndrome (Källstrand et al., 2010), respectively. Bipolar disorder patients have not explicitly been studied before. Two studies with a small number of bipolar patients of which one examined the impact of hallucinatory behavior in mixed groups of psychotic patients reported no significant differences between study groups (Josiassen et al., 1980; McKay et al. 2000). The aim of the current study was to investigate the early auditory processing of complex sound stimuli in BP-I patients and SZ patients using a new technique for recording and analyzing brain evoked responses. Detection of differences in the ABR patterns of BP-I patients compared to those of SZ patients may lead to a better understanding of the etiology and pathophysiological processes underlying BP-I and schizophrenia and determine the validity of the DSM-IV based diagnoses. Furthermore, diagnostic criteria based on a combination of clinical and biological

specifiers could contribute to an earlier correct diagnosis, and hence improvement in treatment. In this study, clinical diagnoses were established using the DSM-IV system, which is solely based on observed and reported signs and symptoms. Several studies have shown, not unexpectedly, a low inter-rater reliability for affective disorder using the DSM-IV, indicating a significant heterogeneity, which however might be possible to diminish, using objective biological measures (Zimmerman et al., 2009). Neurophysiological techniques have shown potential to disclose underlying pathophysiological mechanisms, more or less specific for different psychiatric disorders (reviewed in Onitsuka et al., 2013). Our hypothesis was that an audiometric wave pattern analysis comparing patients with BP-I and SZ would be of interest since this method, using complex sound stimulation and a much more elaborate analysis as compared to standard ABR, has previously identified specific ABR patterns in SZ and Asperger syndrome (Källstrand et al., 2010, 2012). It was thus of primary interest to investigate if specific ABR patterns could be detected also in BP-I, and hence contribute to improved diagnostic precision and also shed some light on the underlying pathohysiological processes in both disorders.

2. Methods 2.1. Subjects 2.1.1. Bipolar disorder patients Thirty outpatients (n ¼30; 16 females, 14 males; mean age 54.67 11.7 years) with a DSM-IV criteria based diagnosis of bipolar disorder type I were recruited from an on-going larger project (The Umeå Bipolar Family Study) aimed at studying the underlying genetic profiles of bipolar disorders. The families were enrolled at the affective unit at Norrlands University Hospital in the northern part of Sweden between 2002 and 2008. In 2009–2010, eligible patients with a well-defined bipolar disorder type I were asked to participate in the ABR substudy. Out of the 30 initially enrolled, 7 (3 females, 4 males) were excluded due to measurement errors. The final group consisted of 23 patients (13 females, 10 males) with a mean age of 54.4 711.7 years at the time of the ABR testing (range 30–72 years). Mean age at onset was 21.2 710.5 years (range 12–53 years) and mean duration of illness 33.27 12.8 years (range 12–52 years). In order to increase the likelihood of similar etiological factors (Faraone et al., 2006), only patients who fulfilled strict criteria for BP-I were included. Also, and a sine qua non for study inclusion, was that the patient had at least one manic episode with psychotic features during the course of the disease. Patients emanated from the same geographical region. Exclusion criteria were beside the usual differential diagnoses (i.e. SZ, schizoaffective disorder, bipolar disorder type II, bipolar disorder UNS, recurrent depression), affective disorders related to CNS pathology including dementia and mental retardation, attention deficit/hyperactivity disorders, borderline personality disorder, alcohol and drug abuse. Furthermore, relatedness, non-Caucasian ancestry, and features that otherwise could compromise the ability to fulfill the study protocol (e.g. not having Swedish as a mother tongue, visual and auditory handicaps) were exclusion criteria. The diagnostic process at entrance in the Umeå Bipolar Family Project was extensive, and repeated at several points in time. It is presented here in an abbreviated version emphasizing the multiple sources of information used. Research psychiatrists and nurses used standard clinical evaluations and research inventories including semi-structured interviews and self-rating scales. We applied standard research inventories as the MINI International Neuropsychiatric

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

Interview (MINI; Sheehan et al., 1998), the Family Interview for Genetic Studies (FIGS) (https://www.nimhgenetics.org/interviews), the Hypomania Check List (HCL-32) (Angst et al., 2005), and shortened version of the Betula neuropsychological assessment inventory (for details, see Nilsson et al., 2004). A somatic examination and information from lifetime medical records was conducted to exclude current somatic disorders, including routine lab, hormoneand metabolic status. The final diagnosis was a best estimate consensus lifetime diagnosis based on DSM-IV (American Psychiatry Association, 2000). Patients signed a written informed consent. The Regional Ethical Review Board at the medical faculty of Umeå University approved the study. Prior to the ABR investigation in 2010, data from the time of inclusion in the Umeå Bipolar Family Study were re-evaluated and a psychiatric and somatic standard examination, including checkup of current medication, confirmation of age at onset, duration of illness, and level of mood (depressed, euthymic, hypomanic/ manic) was performed. All patients except one were in euthymic phase. Lithium solely or in combination with lamotrigine was the most common mood stabilizer (n ¼15), followed by antipsychotic monotherapy (olanzapine, risperidon, clozapine) (n¼ 5) and valproic acid (n ¼ 3). As the bipolar diagnosis was well known to the patient, and no previous results on a bipolar population had been published using this ARB approach, it was explicitly formulated in the informed consent document, that the feed-back to the patients of the results of the study should be limited to a between-group comparison between BP-I and SZ (Källstrand et al., 2012) and a non-psychiatric control population. Furthermore, it was agreed upon that the ABR results should not have any impact on present diagnoses or treatments.

2.1.2. Schizophrenia patients Twenty-three SZ patients (8 females, 15 males; mean age 43.0 711.0 years) were subjected to an ABR test session. Three of these patients (2 females, 1 male) were excluded due to poor data quality, as the audiograms of these patients had poor correlation to an ABR norm curve, indicating that data may be incorrect, possibly due to measurement error. Thus, the final group consisted of twenty patients (7 females, 14 males; mean age 41.37 10.6 years, age range 19–58 years). SZ patients were diagnosed according to DSM-IV by senior psychiatrists. The diagnoses were established at least one year prior to testing. Half of the patients had paranoid SZ, whereas the rest had other subtypes. Eleven patients were at the end of their treatment as in-patients and 12 were out-patients. Mean age at onset was 23.378.3 years (range 17–43 years) and mean duration of illness at time of testing was 17.7710.1 years (range 2–36 years). Patients with co-morbid neuropsychiatric, neurological and significant medical disorders were excluded. All SZ patients had neuroleptic treatment at the time of testing, a majority (80%) was treated with the first generations “high-dose” type neuroleptics. Written informed consent was obtained from all subjects and the study was approved by the ethical committee at the University of Lund, Sweden.

2.1.3. Controls Twenty healthy and normal hearing control individuals were recruited in Lund and Gothenburg, Sweden between 2009 and 2010 and had no known psychiatric diagnoses (10 females, 10 males; mean age 47.4714.0 years; range 22–66 years). Written informed consent was obtained from all controls.

107

2.2. Stimuli and apparatus The evoked potentials were recorded using brainstem audiometer SensoDetects BERA (SensoDetect AB, Lund, Sweden) and using sound stimuli developed by the company. Totally, 13 sound stimuli were used. The sound stimuli included square-shaped click pulses, high frequency varied pulses, low frequency varied pulses, forward masking (see below) and backward masking stimuli. The click pulses were repeated until a total of 1024 accepted evoked potentials had been collected for each sound stimulus. Thus, each ABR waveform represents an average of the responses to 1024 stimulus presentations. TTL trigger pulses coordinated the sweeps with the auditory stimuli. Aberrant activity, such as extremely high amplitudes due to extraordinary movements was rejected. Sound levels were calibrated using a Bruel and Kjaer 2203 sound level meter and Type 4152 artificial ear (Bruel and Kjaer S&V Measurement, Naerum, Denmark). The acoustic output from the earphones corresponds to SPL: 80 dB HL or 109 peSPL (peak equivalence). A square-shaped click pulse was used as probe in the auditory forward masking stimuli (Källstrand et al., 2010). The probe had a duration time of 0.000136 s and a rise and fall time of 0.000023 s. The individual clicks of the stimulus train had an interstimulus interval (ISI) from onset to onset of 0.192 s. In the forward masking paradigm the square-shaped click pulse was preceded by a masker, a 1500 Hz Butterworth low-pass filtered noise. The duration of the masking noise was 0.015 s including the 0.004 s rise and fall time, and the gap between masker and target stimulus was 0.012 s. The time interval onset to onset of click in the forward masking setup was 0.192 s. The masking noise was kept constant at an intensity level of 76 dB SPL. All stimuli were constructed using MATLAB Signal Processing Toolbox (The MathWorks, Inc., Natick, Massachusetts, USA) and stored in a flash memory in the ABR. The stimuli were presented via TDH-50P headphones with Model 51 cushions (Telephonics, Farmingdale, New York, USA). Presentations were made binaurally with the stimuli in phase over headphones. 2.3. Test procedure All tests were performed in a quiet darkened room. Participants were comfortably seated in an armchair in a resting position during the whole testing procedure that lasted approximately 40 min. Surface electrodes were applied to the mastoid bone behind the left and right ear, with one ground electrode placed on the vertex and two reference electrodes placed on the forehead. The subjects were instructed to relax with their eyes closed and were permitted to fall asleep. The test required no active participation other than being subject to sound stimulation. Before the test session, subjects were verbally informed of the nature of the experiments. The click sounds were presented to the subjects beforehand to make them acquainted with the stimuli. 2.4. Data analysis Prior to further analysis the audiogram was correlated with ABR data, from a group of healthy and normal hearing individuals, derived from a normative database to depict general audiogram quality. This is a standard operating procedure of this method in order to grant audiogram quality. A low correlation led to exclusion of the patient due to risk of erroneous measurement (e.g. loose electrodes or head phones). Each individual's forward masking ABR (ABRFM) was imported to Microsoft Excel (Microsoft Corp, Redmond, WA, USA) and analyzed using SensoDetects BAS. Specifically, wave amplitudes and correlation coefficients to a norm ABR curve, respectively, were investigated. Amplitudes were measured from the positive

108

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

peak of a given wave to the bottom of the previous trough. Since the amplitude values obtained were not read in μV, all amplitude values were divided with the highest observed amplitude. Thus, relative linear amplitudes are used in this study. Wave I and II were not analyzed since these peaks were absent in many audiograms. Thus, wave III, V (as waves IV and V often are fused into a single peak), VI and VII amplitudes were investigated. All study participants ABRFM curves were correlated with a median norm ABR curve derived from a group of healthy and normal hearing individuals (n ¼20), obtained from a normative database (see Fig. 1). Three curve regions in the ABRs of BP-I and SZ patients were analyzed and these included waves II–V, waves VI–VII and the whole curve i.e. waves II–VII, respectively (see Fig. 1). These intervals ranged from 2.3 ms to 6.8 ms, 6.8 ms to 11.3 ms and 2.3 ms to 11.3 ms, respectively. Thus, the first interval represents peripheral brainstem and midbrain activity, the second later thalamo-cortical activity and the third the whole ABR pattern. The Spearman rho values were calculated for all regions (rII–V, rVI–VII and rII–VII).

Spearman rho coefficient

Fig. 1. Norm curve showing regions used for calculation of Spearman correlation coefficients.

0.5

***

0.4 0.3 0.2 0.1 0.0 BP-I

SZ

Fig. 2. Correlation between BP-I patient ABR curves (n¼ 23) and norm ABR curve (median curve, n¼20), and between SZ patient ABR curves (n ¼20) and norm ABR curve in region 6.8–11.3 ms of the audiograms (Spearman rho coefficient). Data are mean7 S.E. nnnP o0.001.

(females and males) (Mean-difference¼0.317; p¼0.0004, Mann– Whitney; p¼0.0006, Fischer's exact test; 95% power for po0.05).

3. Results

4. Discussion

When ABR curve patterns upon forward masking stimulation of SZ and BP-I patients were compared to a norm ABR curve derived from healthy controls, that of the BP-I patient group showed a low correlation with the norm curve (see Fig. 2) in the 6.8–11.3 ms region, (corresponding to wave VI–VII), but not in the 2.3 ms to 6.8 ms region nor with the whole ABR curve region (data not shown). Furthermore, for BP-I patients the Spearman rho correlation coefficient rVI-VII was significantly lower than for SZ patients (p ¼0.0003, Mann-Whitney). BP-I and SZ patients did not significantly differ in Spearman correlation coefficients in the other regions i.e. rII–V and rII–VII (data not shown). A majority of BP-I patients had rVI–VII values close to zero, whereas only three SZ patients had a Spearman correlation coefficient rVI–VII o0.1. Thereafter, ABR activity was investigated in waves III, V, VI and VII. No significant differences in peak amplitudes were found except for wave VII that was significantly higher in the BP-I group compared to the SZ group, Mann–Whitney P ¼0.0062, and also, wave III amplitude was significantly higher in the BP-I patient group, Mann–Whitney P ¼0.0472 (see Fig. 3). Males and females analyzed separately showed results in the same direction. For the strongest correlation trait r(VI–VII) BP1 females show lower relative amplitudes than SZ females (Mean-difference¼ 0.327; p ¼0.0159, Mann–Whitney; p ¼0.0050, Fischer's exact test; 85% power for p o 0.05). Likewise, BP1 males show lower relative amplitudes than SZ males (Meandifference ¼0.286; p ¼ 0.0547, Mann–Whitney; p ¼0.0349, Fischer's exact test; 60% power for p o 0.05). Finally, BP1 (females and males) show lower relative amplitudes than SZ

This study shows that both BP-I, and SZ patients to some extent, have an influence on the auditory pathway as measured by the ABR technique. Interestingly, both disorders, and in particular BP-I, differed from the ABR norm curve which was based on healthy individuals. Abnormalities in the auditory brainstem response in SZ and Asperger disorder upon complex sound stimulation have previously been shown (Källstrand et al., 2010; Källstrand et al., 2012). BP-I patients differed from the norm curve in a region of the curve that is related to thalamus and thalamocortical structures. In this region BP-I patients also showed increased activity compared with SZ patients, which has been observed previously (see below). BP-I patients also differed from SZ patients earlier in the auditory pathway i.e. at the level of cochleus nucleus. More specifically, the current study shows an increased activity in wave VII and wave III upon forward masking stimuli in BP-I patients. It was also found that BP-I patients had low correlation to a norm ABR curve in the region comprising wave VI–VII. Both SZ and BP-I patients differed from the norm curve in this region, although BP-I patients contrasted to a higher degree. The BP-I patients also differed significantly from the SZ patients. The increased activity in wave VII may partially contribute to this finding, however, it can be assumed that further aberrances in this region are also reflected. The notable difference between BP-I patients and norm curve is likely to reflect abnormalities in addition to increased wave VII amplitude which would be interesting to study further. Thus, the main results of this study are found in the region comprising wave VI–VII, which has been suggested to reflect activity in the thalamic region, specifically the medial geniculate body, although other structures e.g. contralateral

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

*

0.3 0.2 0.1 0.0

0.2

0.1

0.0

BP-I

SZ

BP-I

ns

0.15 0.10 0.05

SZ

**

0.25

relative amplitude

0.20

relative amplitude

ns

0.3

relative amplitude

relative amplitude

0.4

109

0.20 0.15 0.10 0.05 0.00

0.00 BP-I

SZ

BP-I

SZ

Fig. 3. Mean relative amplitudes for wave III (A), wave V (B), wave VI (C) and wave VII (D) for BP-I patients (n¼23) and SZ patients (n¼ 20), respectively. Data are mean 7 S.E. n P o0.05, nnPo 0.01.

inferior colliculus and thalamocortical networks, may also be reflected (Burkard et al., 2007). Thalamus is a key structure within the corticolimbic circuitry that modulates mood states and is also involved in various cognitive processes which can be impaired in bipolar disorder (Sax et al., 1999). Thalamus also plays a prominent role in processing and integrating incoming sensory information, e.g. by inhibitory processes modulating afferent information (Steriade, 2004) and signals between subcortical and cortical regions as well as links different cortical regions via the cortico-thalamo-cortical pathways. Most volumetric studies have shown that the thalamus volume relative to whole brain is decreased in SZ (reviewed in Byne et al., 2009), whereas MRI studies of thalamic volume have shown various results in bipolar disorder (Dasari et al., 1999;Dupont et al., 1995; Radenbach et al., 2010; Sax et al., 1999; Strakowski et al., 1993, 1999). Interestingly, one study showed increased volume of the lateral geniculate nucleus of the thalamus, the visual relay station of the thalamus, in mood disorders but not in SZ (Dorph-Peterson et al., 2009). Functional abnormalities in the thalamus have also been reported such as abnormal metabolism in medial thalamus in bipolar depression (Buchsbaum et al., 1997) and increased thalamic creatine, N-acetylaspartate (Deicken et al., 2001) and choline metabolites in bilateral thalami (Howells et al., 2013) of euthymic BP-I patients. In contrast, thalamic N-acetylaspartate levels are reduced in SZ (Deicken et al., 2000). In line with the results of the current investigation showing increased activation at the level of thalamus, a recent metaanalysis investigating emotional processing using fMRI showed overactivity upon emotional facial stimuli in subcortical regions such as the thalamus but not cortical regions in bipolar disorder patients compared to schizophrenic patients (Delvecchio et al., 2013). Thus, various forms of thalamic pathology have been implicated in bipolar disorders. Regarding our specific findings suggesting aberrant functioning of at the level of thalamus, the contralateral inferior colliculus and/or thalamocortical networks, the cause of this

may be difficult to pinpoint. The inferior colliculus receives auditory information from the lower brainstem on amplitude, frequency and lateralization as well as general brainstem sensory and motor information, and projects to a large extent to the medial geniculate body, the principal auditory nucleus of the thalamus. In the inferior colliculus the signals may be modified, sharpened or de novo response properties may be formed (Pollak et al., 2011). Thalamus is a central processor to direct incoming and outgoing signals from telencephalic and lower structures. There is facilitation and inhibition from both directions. The enhanced activity of wave VII may have arisen by disturbed inhibitory function from central deficiencies or by a specific pathology of the thalamus itself, which could indicate inadequate inhibitory processing. Furthermore, the small but significant increase of wave III amplitude (reflecting activity at the level of the cochlear nucleus) in the BP-I group may have effect on the enhanced activity in wave VII since there are direct interconnections at all levels in the auditory pathway. Abnormalities in the early auditory processing at the level of the cochlear nucleus, have not been thoroughly studied in bipolar disorder. Such abnormalities associated with bipolar disorder is thus a novel finding. Interestingly, acoustic neuromas have been associated with manic and mixed states (Kalayam et al., 1994). Many investigations such as neuroimaging, neuropathological, auditory event-related potentials and genetic studies have to some extent found similar abnormalities in bipolar disorder and schizophrenia. A recent genetic study investigating genome-wide SNPs showed a high genetic correlation between schizophrenia and bipolar disorder (Cross-Disorder Group of the Psychiatric Genomics Consortium, 2013), in agreement with family studies (Cardno et al., 2002; Lichtenstein et al., 2009; Mortensen et al., 2003) although there are also contradicting results suggesting no genetic overlap between schizophrenia and bipolar disorder (Kendler et al., 1993; Maier et al., 1993). The ABR technique used in the present study also showed some overlap although the differences between

110

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

the disorders were found to be more pronounced and, furthermore, revealed specific ABR patterns in BP-I. Today, there are no biomarkers that separate BP-I and SZ with high sensitivity and specificity. The findings in this study, however, show that this ABR technique using forward masking stimuli, efficiently separates BP-I from SZ. Furthermore, it also convincingly demonstrates that the ABR pattern observed in BP-I, may be a strong candidate for serving as a biomarker for this disorder. The increased wave VII amplitude separated BP-I patients from SZ patients with 65% sensitivity and 80% specificity whereas the increased wave III amplitude identified BP-I with 56% sensitivity and 80% specificity (data not shown). However, further studies are needed to confirm these results since the sample sizes in this study were relatively small and not matched for age and gender. Furthermore, longitudinal studies are needed to investigate if the observed ABR abnormalities in BP-I are present already in the early stages of the disorder or are affected by medication. In line with the clinical as well as the etiopathological heterogeneity within the BP-I and SZ groups respectively, it was however not surprising that the results showed some degree of overlapping between the disorders. Fifty per cent of the SZ patients had correlation rVI–VII 4 0.4 with the norm curve whereas only two BP-I patients had rVI–VII 40.4. Thus most patients, in particular the BP-I patients, had low correlation with the norm curve. Furthermore, a majority of the BP-I patients (60.9%) had very low correlation (rVI–VII o0.03) with the norm curve whereas 95% of the SZ patients had a higher correlation (rVI–VII 40.03), efficiently separating BP-I from SZ (data not shown). Thus, in this study the ABR curves of the BP-I patients are clearly separated from controls (i.e. the norm curve) whereas some overlap is observed between SZ patients and controls. Furthermore, increased wave III and VII amplitudes distinguish BP-I patients from SZ patients. The most interesting and robust findings in this study are the highly significant abnormalities found in the region comprising wave VI–VII in BP-I; the poor correlation to the norm curve in this region and the increased wave VII amplitude. It would be of great interest to further study this region in BP-I, and to include other indications such as attention deficit/hyperactivity disorders, borderline personality disorder and other affective disorders in future research.

5. Conclusion The current study identified deficits in the auditory pathway in the brainstem that is specific for BP-I and not shared with schizophrenia. Specifically, the region comprising waves VI–VII was significantly different from that of healthy individuals and schizophrenia patients, which could partly be explained by the finding of increased wave VII amplitude in BP-I patients. Thus, abnormalities in the region comprising waves VI–VII is a prominent finding in BP-I and may serve as a biological marker for BP-I. Although the results presented here can already be of clinical relevance in specialized affective disorders centra, larger studies of well-characterized patients will be needed to confirm these results.

Role of funding source Provided funding for The Umeå Bipolar Family Study aimed at studying the underlying genetic profiles of bipolar disorders.

Conflict of interest Authors Johan Källstrand and Sara Nehlstedt are employees of, and hold stock in SensoDetect AB. Authors Mia Sköld and Jens Holmberg are employees of SensoDetect AB. Author Sören Nielzén hold stock in SensoDetect AB. Authors Rolf Adolfsson and Annelie Nordin declare that they have no conflicts of interest.

Acknowledgment The authors wish to thank the Västerbotten County Councils for funding the present project.

References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association, Washington, DC (text revision). Angst, J., Adolfsson, R., Benazzi, F., Gamma, A., Hantouche, E., Meyer, T.D., Skeppar, P., Vieta, E., Scott, J., 2005. The HCL-32: towards a self-assessment tool for hypomanic symptoms in outpatients. J. Affect. Disord. 88, 217–233. Baldessarini, R.J., Tondo, L., Baethge, C.J., Lepri, B., Bratti, I.M., 2007. Effects of treatment latency on response to maintenance treatment in manic-depressive disorder. Bipolar Disord. 9, 386–393. Bowden, C.L., 2001. Strategies to reduce misdiagnosis of bipolar depression. Psychiatr. Serv. 52, 51–55. Bramon, E., Rabe-Hesketh, S., Sham, P., Murray, R.M., Frangou, S., 2004. Metaanalysis of the P300 and P50 waveforms in schizophrenia. Schizophr. Res. 70, 315–329. Brecher, M., Begleiter, H., 1985. Brain stem auditory evoked potentials in unmedicated schizophrenic patients. Biol. Psychiatry 20, 199–202. Buchsbaum, M.S., Wu, J., Siegel, B.V., Hackett, E., Trenary, M., Abel, L., Reynolds, C., 1997. Effects of sertraline on regional metabolic rate in patients with affective disorder. Biol. Psychiatry 41, 15–22. Burkard, R.F., Don, M., Eggermont, J.J., 2007. Auditory Evoked Potentials: Basic Principles and Clinical Application. Lippincott Williams and Wilkins, Baltimore, MD. Byne, W., Hazlett, E.A., Buchsbaum, M.S., Kemether, E., 2009. The thalamus and schizophrenia: current status of research. Acta Neuropathol. 117, 347–368. Cardno, A.G., Rijsdijk, F.V., Sham, P.C., Murray, R.M., McGuffin, P., 2002. A twin study of genetic relationships between psychotic symptoms. Am. J. Psychiatry 159, 539–545. Catts, S.V., Shelley, A.M., Ward, P.B., Liebert, B., McConaghy, N., Andrews, S., Michie, P.T., 1995. Brain potential evidence for an auditory sensory memory deficit in schizophrenia. Am. J. Psychiatry 152, 213–219. Chiappa, K.H., 1997. Evoked Potentials in Clinical Medicine. Lippincott-Raven Publishers, Philadelphia, PA. Cross-Disorder Group of the Psychiatric Genomics Consortium, Lee, S.H., Ripke, S., Neale, B.M., Faraone, S.V., Purcell, S.M., Perlis, R.H., Mowry, B.J., Thapar, A., Goddard, M.E., Witte, J.S., Absher, D., Agartz, I., Akil, H., Amin, F., Andreassen, O.A., Anjorin, A., Anney, R., Anttila, V., Arking, D.E., Asherson, P., Azevedo, M.H., Backlund, L., Badner, J.A., Bailey, A.J., Banaschewski, T., Barchas, J.D., Barnes, M.R., Barrett, T.B., Bass, N., Battaglia, A., Bauer, M., Bayés, M., Bellivier, F., Bergen, S.E., Berrettini, W., Betancur, C., Bettecken, T., Biederman, J., Binder, E.B., Black, D.W., Blackwood, D.H., Bloss, C.S., Boehnke, M., Boomsma, D.I., Breen, G., Breuer, R., Bruggeman, R., Cormican, P., Buccola, N.G., Buitelaar, J.K., Bunney, W.E., Buxbaum, J.D., Byerley, W.F., Byrne, E.M., Caesar, S., Cahn, W., Cantor, R.M., Casas, M., Chakravarti, A., Chambert, K., Choudhury, K., Cichon, S., Cloninger, C.R., Collier, D.A., Cook, E.H., Coon, H., Cormand, B., Corvin, A., Coryell, W.H., Craig, D.W., Craig, I.W., Crosbie, J., Cuccaro, M.L., Curtis, D., Czamara, D., Datta, S., Dawson, G., Day, R., De Geus, E.J., Degenhardt, F., Djurovic, S., Donohoe, G.J., Doyle, A.E., Duan, J., Dudbridge, F., Duketis, E., Ebstein, R.P., Edenberg, H.J., Elia, J., Ennis, S., Etain, B., Fanous, A., Farmer, A.E., Ferrier, I.N., Flickinger, M., Fombonne, E., Foroud, T., Frank, J., Franke, B., Fraser, C., Freedman, R., Freimer, N.B., Freitag, C.M., Friedl, M., Frisén, L., Gallagher, L., Gejman, P.V., Georgieva, L., Gershon, E.S., Geschwind, D.H., Giegling, I., Gill, M., Gordon, S.D., Gordon-Smith, K., Green, E.K., Greenwood, T.A., Grice, D.E., Gross, M., Grozeva, D., Guan, W., Gurling, H., De Haan, L., Haines, J.L., Hakonarson, H., Hallmayer, J., Hamilton, S.P., Hamshere, M.L., Hansen, T.F., Hartmann, A.M., Hautzinger, M., Heath, A.C., Henders, A.K., Herms, S., Hickie, I.B., Hipolito, M., Hoefels, S., Holmans, P.A., Holsboer, F., Hoogendijk, W.J., Hottenga, J.J., Hultman, C.M., Hus, V., Ingason, A., Ising, M., Jamain, S., Jones, E.G., Jones, I., Jones, L., Tzeng, J.Y., Kähler, A.K., Kahn, R.S., Kandaswamy, R., Keller, M.C., Kennedy, J.L., Kenny, E., Kent, L., Kim, Y., Kirov, G.K., Klauck, S.M., Klei, L., Knowles, J.A., Kohli, M.A., Koller, D.L., Konte, B., Korszun, A., Krabbendam, L., Krasucki, R., Kuntsi, J., Kwan, P., Landén, M., Långström, N., Lathrop, M., Lawrence, J., Lawson, W.B., Leboyer, M., Ledbetter, D.H., Lee, P.H., Lencz, T., Lesch, K.P., Levinson, D.F., Lewis, C.M., Li, J., Lichtenstein, P., Lieberman, J.A., Lin, D.Y., Linszen, D.H., Liu, C., Lohoff, F.W., Loo, S.K., Lord, C., Lowe, J.K., Lucae, S., MacIntyre, D.J., Madden, P.A., Maestrini, E., Magnusson, P.K., Mahon, P.B., Maier, W., Malhotra, A.K., Mane, S.M., Martin, C.L., Martin, N.G., Mattheisen, M., Matthews, K., Mattingsdal, M., McCarroll, S.A., McGhee, K.A., McGough, J.J., McGrath, P.J., McGuffin, P., McInnis, M.G., McIntosh, A., McKinney, R., McLean, A.W., McMahon, F.J., McMahon, W.M., McQuillin, A., Medeiros, H., Medland, S.E., Meier, S., Melle, I., Meng, F., Meyer, J., Middeldorp, C.M., Middleton, L., Milanova, V., Miranda, A., Monaco, A.P., Montgomery, G.W., Moran, J.L., Moreno-De-Luca, D., Morken, G., Morris, D.W., Morrow, E.M., Moskvina, V., Muglia, P., Mühleisen, T.W., Muir, W.J., Müller-Myhsok, B., Murtha, M., Myers, R.M., Myin-Germeys, I., Neale, M.C., Nelson, S. F., Nievergelt, C.M., Nikolov, I., Nimgaonkar, V., Nolen, W.A., Nöthen, M.M., Nurnberger, J.I., Nwulia, E.A., Nyholt, D.R., O’Dushlaine, C., Oades, R.D., Olincy, A., Oliveira, G., Olsen, L., Ophoff, R.A., Osby, U., Owen, M.J., Palotie, A., Parr, J.R., Paterson, A.D., Pato, C.N., Pato, M.T., Penninx, B.W., Pergadia, M.L., Pericak-Vance, M.A., Pickard, B.S., Pimm, J., Piven, J., Posthuma, D., Potash, J.B., Poustka, F., Propping, P., Puri, V., Quested, D.J., Quinn, E.M., Ramos-Quiroga, J.A., Rasmussen, H.B., Raychaudhuri, S., Rehnström, K., Reif, A., Ribasés, M., Rice, J.P., Rietschel, M., Roeder, K., Roeyers, H., Rossin, L., Rothenberger, A., Rouleau, G., Ruderfer, D., Rujescu, D., Sanders, A.R.,

M. Sköld et al. / Journal of Affective Disorders 169 (2014) 105–111

Sanders, J., Santangelo, S.L., Sergeant, J.A., Schachar, R., Schalling, M., Schatzberg, A.F., Scheftner, W.A., Schellenberg, G.D., Scherer, S.W., Schork, N.J., Schulze, T.G., Schumacher, J., Schwarz, M., Scolnick, E., Scott, L.J., Shi, J., Shilling, P.D., Shyn, S.I., Silverman, J.M., Slager, S.L., Smalley, S.L., Smit, J.H., Smith, E.N., Sonuga-Barke, E.J., St Clair, D., State, M., Steffens, M., Steinhausen, H.C., Strauss, J.S., Strohmaier, J., Stroup, T. S., Sutcliffe, J.S., Szatmari, P., Szelinger, S., Thirumalai, S., Thompson, R.C., Todorov, A. A., Tozzi, F., Treutlein, J., Uhr, M., van den Oord, E.J., Van Grootheest, G., Van Os, J., Vicente, A.M., Vieland, V.J., Vincent, J.B., Visscher, P.M., Walsh, C.A., Wassink, T.H., Watson, S.J., Weissman, M.M., Werge, T., Wienker, T.F., Wijsman, E.M., Willemsen, G., Williams, N., Willsey, A.J., Witt, S.H., Xu, W., Young, A.H., Yu, T.W., Zammit, S., Zandi, P.P., Zhang, P., Zitman, F.G., Zöllner, S., International Inflammatory Bowel Disease Genetics Consortium (IIBDGC), Devlin, B., Kelsoe, J.R., Sklar, P., Daly, M.J., O’Donovan, M.C., Craddock, N., Sullivan, P.F., Smoller, J.W., Kendler, K.S., Wray, N.R., 2013. Genetic relationship between five psychiatric disorders estimated from genome-wide SNPs. Nat. Genet. 45, 984–995. Dasari, M., Friedman, L., Jesberger, J., Stuve, T.A., Findling, R.L., Swales, T.P., Schulz, S.C., 1999. A magnetic resonance imaging study of thalamic area in adolescent patients with either schizophrenia or bipolar disorder as compared to healthy controls. Psychiatry Res. 91, 155–162. Deicken, R.F., Johnson, C., Eliaz, Y., Schuff, N., 2000. Reduced concentrations of thalamic N-acetylaspartate in male patients with schizophrenia. Am. J. Psychiatry 157, 644–647. Deicken, R.F., Eliaz, Y., Feiwell, R., Schuff, N., 2001. Increased thalamic N-acetylaspartate in male patients with familial bipolar I disorder. Psychiatry Res. 106, 35–45. Delvecchio, G., Sugranyes, G., Frangou, S., 2013. Evidence of diagnostic specificity in the neural correlates of facial affect processing in bipolar disorder and schizophrenia: a meta-analysis of functional imaging studies. Psychol. Med. 43, 553–569. Dorph-Peterson, K.-A., Caric, D., Saghafi, R., Zhang, W., Sampson, A.R., Lewis, D.A., 2009. Volume and neuron number of lateral geniculate nucleus in schizophrenia and mood disorders. Acta Neuropathol 117, 369–384. Dupont, R.M., Jernigan, T.L., Heindel, W., Butters, N., Shafer, K., Wilson, T., Hesselink, J., Gillin, J.C., 1995. Magnetic resonance imaging and mood disorders. Arch. Gen. Psychiatry 52, 747–755. Faraone, S.V., Lasky-Su, J., Glatt, S.J., Van Eerdewegh, P., Tsuang, M.T., 2006. Early onset bipolar disorder: possible linkage to chromosome 9q34. Bipolar Disord. 8, 144–151. Grillon, C., Ameli, R., Glazer, W.M., 1990. Brainstem auditory-evoked potentials to different rates and intensities of stimulation in schizophrenics. Biol. Psychiatry 28, 819–823. Hall, M.H., Schulze, K., Rijsdijk, F., Kalidindi, S., McDonald, C., Bramon, E., Murray, R.M., Sham, P., 2009. Are auditory P300 and duration MMN heritable and putative endophenotypes of psychotic bipolar disorder? A Maudsley bipolar twin and family Study. Psychol. Med. 39, 1277–1287. Hauser, M., Pfenning, A., Özgürdal, S., Heinz, A., Bauer, M., Juckel, G., 2007. Early recognition of bipolar disorder. Eur. Psychiatry 22, 92–98. Hayashida, Y., Mitani, Y., Hosomi, H., Amemiya, M., Mifune, K., Tomita, S., 1986. Auditory brain stem responses in relation to the clinical symptoms of schizophrenia. Biol. Psychiatry 21, 177–188. Howells, F.M., Ives-Deliperi, V.L., Horn, N.R., Stein, D.J., 2013. Increased thalanmic phospholipid concentration evident in bipolar I disorder. Prog. NeuroPsychopharmacol. Biol. Psychiatry 41, 1–5. Igata, M., Ohta, M., Hayashida, Y., Abe, K., 1994. Missing peaks in auditory brainstem responses and negative symptoms in schizophrenia. Jpn. J. Psychiatry Neurol. 48, 571–578. Josiassen, R.C., Busk, J., Hart, A.D., Vanderploeg, R., 1980. Early auditory information processing in schizophrenia: a preliminary report. Biol. Psychol. 10, 225–234. Kalayam, B., Young, R.C., Tsuboyama, G.K., 1994. Mood disorders associated with acoustic neuromas. Int. J. Psychiatry Med. 24, 31–43. Källstrand, J., Olsson, O., Nehlstedt, S.F., Sköld, M.L., Nielzén, S., 2010. Abnormal auditory forward masking pattern in the brainstem response of individuals with Asperger syndrome. Neuropsychiatr. Dis. Treat. 6, 289–296. Källstrand, J., Nehlstedt, S.F., Sköld, M.L., Nielzén, S., 2012. Lateral asymmetry and reduced forward masking effect in early brainstem auditory evoked responses in schizophrenia. Psychiatry Res. 196, 188–193. Kendell, R., Jablensky, A., 2003. Distinguishing between the validity and utility of psychiatric diagnoses. Am. J. Psychiatry 160, 4–12. Kendler, K.S., McGuire, M., Gruenberg, A.M., O’Hare, A., Spellman, M., Walsh, D., 1993. The Roscommon family study. IV. Affective illness, anxiety disorders, and alcoholism in relatives. Archiv. Gen. Psychiatry 50, 952–960. Lichtenstein, P., Yip, B.H., Björk, C., Pawitan, Y., Cannon, T.D., Sullivan, P.F., Hultman, C.M., 2009. Common genetic determinants of schizophrenia and bipolar disorder in Swedish families: a population-based study. Lancet 373, 234–239.

111

Lindström, L., Klockhoff, I., Svedberg, A., Bergström, K., 1987. Abnormal auditory brain-stem responses in hallucinating schizophrenic patients. Lateral asymmetry and reduced forward masking effect in early brainstem auditory evoked responses in schizophrenia. Psychiatry Res. 196, 188–193. Lindström, L.H., Wieselgren, I-M., Klockhoff, I., Svedberg, A., 1990. Relationship between abnormal brainstem auditory-evoked potentials and subnormal CSF levels of HVA and 5-HIAA in first-episode schizophrenic patients. Biol. Psychiatry 28, 435–442. Maier, W., Lichtermann, D., Minges, J., Hallmayer, J., Heun, R., Benkert, O., Levinson, D.F., 1993. Continuity and discontinuity of affective disorders and schizophrenia. Results of a controlled family study. Arch. Gen. Psychiatry 50, 871–883. McKay, C.M., Headlam, D.M., Copolov, D.L., 2000. Central auditory processing in patients with auditory hallucinations. Am. J. Psychiatry 157, 759–766. Mitchell, P.B., 2012. Bipolar disorder: the shift to overdiagnosis. Can. J. Psychiatry 57, 659–665. Mortensen, P.B., Pedersen, C.B., Melbye, M., Mors, O., Ewald, H., 2003. Individual and familial risk factors for bipolar affective disorders in Denmark. Arch. Gen. Psychiatry 60, 1209–1215. Murray, R.M., Sham, P., Van Os, J., Zanelli, J., Channon, M., McDonald, C., 2004. A developmental model for similarities and dissimilarities between schizophrenia and bipolar disorder. Schizophr. Res. 71, 405–416. Ng, W.X., Lau, I.Y., Graham, S., Sim, K., 2009. Neurobiological evidence for thalamic, hippocampal and related glutamatergic abnormalities in bipolar disorder: a review and synthesis. Neurosci. Biobehav. Rev. 33, 336–354. Nilsson, L.-G., Adolfsson, R., Bäckman, L., de Frias, C.M., Molander, B., Nyberg, L., 2004. Betula: a prospective cohort study on memory, health and aging. Aging, Neuropsychol. Cognit. 11, 134–148. Olincy, A., Martin, L., 2005. Diminished suppression of the P50 auditory evoked potential in bipolar disorder subjects with a history of psychosis. Am. J. Psychiatry 162, 43–49. Onitsuka, T., Oribe, N., Kanba, S., 2013. Neurophysiological findings in patients with bipolar disorder. Suppl. Clin. Neurophysiol. 62, 197–206. Pfefferbaum, A., Horvath, T.B., Roth, W.T., Tinklenberg, J.R., Kopell, B.S., 1980. Auditory brain stem and cortical evoked potentials in schizophrenia. Biol Psychiatry 15, 209–223. Pollak, G.D., Xie, R., Gittelman, J.X., Andoni, S., Li, N., 2011. The dominance of inhibition in the inferior colliculus. Hear. Res. 274, 27–39. Post, R.M., Leverich, G.S., Kupka, R.W., Keck Jr., P.E., McElroy, S.L., Altshuler, L.L., Frye, M.A., Luckenbaugh, D.A., Rowe, M., Grunze, H., Suppes, T., Nolen, W.A., 2010. Early-onset bipolar disorder and treatment delay are risk factors for poor outcome in adulthood. J. Clin. Psychiatry 71, 864–872. Radenbach, K., Flaig, V., Schneider-Axmann, T., Usher, J., Reith, W., Falkai, P., Gruber, O., Scherk, H., 2010. Thalamic volumes in patients with bipolar disorder. Eur. Arch. Psychiatry Clin. Neurosci. 260, 601–607. Sax, K.W., Strakowski, S.M., Zimmermann, M.E., DelBello, M.P., Keck Jr., P.E., Hawkins, J.M., 1999. Frontosubcortical neuroanatomy and the continuous performance test in mania. Am. J. Psychiatry 156, 139–141. Schulze, K.K., Hall, M.H., McDonald, C., Marshall, N., Walshe, M., Murray, R.M., Bramon, E., 2007. P50 auditory evoked potential suppression in bipolar disorder patients with psychotic features and their unaffected relatives. Biol. Psychiatry 62, 121–128. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M. I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 9, 22–33. Steriade, M., 2004. Local gating of information processing through the thalamus. Neuron 41, 493–494. Strakowski, S.M., DelBello, M.P., Sax, K.W., Zimmerman, M.E., Shear, P.K., Hawkins, J.M., Larson, E.R., 1999. Brain magnetic resonance imaging of structural abnormalities in bipolar disorder. Arch. Gen. Psychiatry 56, 254–260. Strakowski, S.M., Delbello, M.P., Adler, C.M., 2005. The functional neuroanatomy of bipolar disorder: a review of neuroimaging findings. Mol. Psychiatry 10, 105–116. Strakowski, S.M., Wilson, D.R., Tohen, M., Woods, B.T., Douglass, A.W., Stoll, A.L., 1993. Structural brain abnormalities in first-episode mania. Biol. Psychiatry 33, 602–609. Swann, A.C., Bowden, C.L., Calabrese, J.R., Dilsaver, S.C., Morris, D.D., 1999. Differential effect of number of previous episodes of affective disorder on response to lithium or divalproex in acute mania. Am. J. Psychiatry 156, 1264–1266. Umbricht, D., Koller, R., Schmid, L., Skrabo, A., Grübel, C., Huber, T., Stassen, H., 2003. How specific are deficits in mismatch negativity generation to schizophrenia? Biol. Psychiatry 53, 1120–1131. Zimmerman, P., Brücki, T., Nocon, A., Pfister, H., Lieb, R., Wittchen, H.U., Holsboer, F., Angst, J., 2009. Heterogeneity of DSM-IV major depressive disorder as a consequence of subtreshold bipolarity. Arch. Gen. Psychiatry 66, 1341–1352.