- Email: [email protected]
Correlation of brain default mode network activation with bipolarity index in youth with mood disorders
Journal of Affective Disorders 150 (2013) 1174–1178
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Preliminary communication
Correlation of brain default mode network activation with bipolarity index in youth with mood disorders Kristen A. Ford a, Jean Théberge b, Richard J. Neufeld a,c,d, Peter C. Williamson a,b, Elizabeth A. Osuch a,b,n a
University of Western Ontario, Schulich School of Medicine and Dentistry, Department of Psychiatry, London, Ontario, Canada University of Western Ontario, Department of Medical Biophysics, and Lawson Health Research Institute, London, Ontario, Canada c University of Western Ontario, Department of Psychology, London, Ontario, Canada d University of Western Ontario, Program in Neuroscience, London, Ontario, Canada b
art ic l e i nf o
a b s t r a c t
Article history: Received 27 March 2013 Received in revised form 22 May 2013 Accepted 31 May 2013 Available online 24 June 2013
Background: Major Depressive Disorder (MDD) and Bipolar Disorder (BD) can be difficult to differentiate, as both feature depressive episodes. Here we have utilized fMRI and a measure of trait bipolarity to examine resting-state functional connectivity of brain activation in the default mode network in youth with MDD and BD to isolate trait-specific patterns. Methods: We collected resting-state fMRI scans from thirty youth (15 MDD; 15 BD, Type 1). The Bipolarity Index (BI) was completed by each patient's treating psychiatrist. Independent components analysis was used to extract a default mode network component from each participant, and then multiple regression was used to identify correlations between bipolarity and network activation. Results: Activation in putamen/claustrum/insula correlated positively with BI; activation in the postcentral gyrus/posterior cingulate gyrus correlated negatively with BI. These correlations did not appear to be driven by movement in the scanner, state depression, gender or lithium use. Limitations: There were group differences in state depression and sex that needed to be statistically covaried; differences in medication use existed between the groups; sample size was not large. Conclusions: The identification of the putamen/claustrum in our positive correlation may indicate a potential trait marker for the psychomotor activation unique to bipolar mania. The negative correlation in the postcentral gyrus/posterior cingulate suggests that this functional inactivation is more specific to MDD and is consistent with previous research. Ultimately, this approach may help to develop techniques to minimize the current clinical dilemma by facilitating the classification between BD and MDD. & 2013 Elsevier B.V. All rights reserved.
Keywords: Bipolar Disorder Youth mental health Major Depressive Disorder Functional brain imaging Bipolar spectrum Default mode network
1. Introduction Major Depressive Disorder (MDD) and Bipolar Disorder (BD) both feature depressive episodes, with depression the most common presenting complaint for both in ambulatory settings. BD patients go an average of 6–10 years without the proper diagnosis, usually diagnosed with MDD (Berk et al., 2010). Correct identification is critical because antidepressant use in patients with BD can lead to worsening of symptoms (Berk et al., 2010; Shi et al., 2004). The question has been raised of whether affective illnesses lie along a continuum from MDD to BD. The construct of “bipolar spectrum disorders” was developed to account for patients without mania or hypomania who had mood cycling (Akiskal et al., 1977). Certainly evidence showing that half of patients originally meeting criteria for
n Correspondence to: Schulich School of Medicine and Dentistry, Department of Psychiatry, 860 Richmond Street, London, ON N6A 3H8; Canada. Tel.: +1 519 646 6000x65188; fax: +1 519 646 6211. E-mail address: [email protected] (E.A. Osuch).
0165-0327/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jad.2013.05.088
MDD converted to a form of BD over three decades of study (Angst et al., 2005) indicates the current challenges of long-term differentiation between these disorders. It has been suggested that, while there appears to be a form of depression that indicates a “pure” MDD (Zimmermann et al., 2009), the rest of the primary affective disorders may lie on a continuum with BD. The purpose of this study was to examine the extent to which a measure of “bipolarity” would correlate with functional network connectivity in youth recently diagnosed with either MDD or BD. We hypothesized that participants with these mood disorders would demonstrate differences in resting-state default mode network (DMN) connectivity that would be related to the construct known as bipolarity, which constitutes a continuous variable measurable across participant groups.
2. Methods Approval for the protocol was obtained from the Research Ethics Board at the University of Western Ontario, Schulich School
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
of Medicine and Dentistry. Participants were recruited from the First Episode Mood and Anxiety Program (FEMAP) in London, Ontario, Canada. After a complete description of the study to participants, written informed consent was obtained. For each participant, Bipolarity Index scores (BI) (Sachs, 2004) were completed by their treating psychiatrist. Three team psychiatrists completed the BI on 27 participants and 3 other psychiatrist did the remainder. The BI provides a score (between 0 and 100) using five “dimensions of bipolarity” including mood symptoms, age of onset of symptoms, course of illness, response to medications, and family history of mood and substance use. The instrument measures trait features, independent of mood state at any given time. Each participant also met diagnostic criteria for Bipolar Disorder, type I or Major Depressive Disorder using the Structured Clinical Interview for Diagnosis-IV. Medications were unchanged for three weeks prior to scanning. The Beck Depression Inventory (BDI)(Beck et al., 1996) and Young Mania Rating Scale (YMRS)(Young et al., 1978) were administered on scanning day. All magnetic resonance imaging (MRI) was performed using a 3.0 T MRI scanner (Siemens Verio, Erlangen, Germany) at the Lawson Health Research Institute, and a 32-channel phased-array head coil (Siemens). Whole-brain, T1-weighted, anatomical images with 1 mm isotropic spatial resolution were acquired as reference for spatial normalization. An anatomical image was used to select the orientation of functional MRI images 61 coronal to the AC–PC plane. Functional scans consisted of gradient-echo, echo-planar scans (TR ¼3 s, TE ¼30 ms, FOV ¼240 mm 240 mm, matrix ¼80 80, flip angle 901, 40 transverse slices, thickness¼ 3 mm) with no parallel acceleration, covering whole brain with an isotropic spatial resolution of 3 mm for a total time of 8 min, 9 s per run (161 brain volumes plus two discarded steadystate volumes). Participants were instructed to lie comfortably with their eyes closed and let their minds wander without falling asleep. All participants reported compliance with these instructions. fMRI data underwent standard pre-processing including motion correction, spatial normalization into the standard Montreal Neurological Institute (MNI) space and spatial smoothing with a 3D-Gaussian kernel (10 mm FWHM radius) in preparation for statistical analysis (SPM8). Whole-brain resting-state fMRI time courses of all participants were entered into a group spatial Independent Component Analyses (ICA). Using Group ICA of fMRI Toolbox (GIFT) software (http://mialab.mrn.org/software/gift/ index.html), this procedure decomposes the group data into maximally independent functional brain components by dimension estimation and data reduction with Principal Component Analysis followed by independent component estimation using the Infomax algorithm. This ICA analysis produced a series of 30 temporal components for each participant. Individual spatial maps were then constructed for each participant and converted to Z-scores. These represented the relative strengths of each component identified in the data of each participant. The DMN was identified as follows. Individual participant components were spatially sorted based on the strength of the correlation between each component and a standardized map of the DMN that was constructed to include Brodmann's areas 7, 10, 39, the precuneus, and the posterior cingulate. For each participant, the component with the highest correlation value to this map was selected as the DMN component-of-interest. A whole brain multiple regression analysis (SPM8) was conducted which incorporated three covariates (BDI score, BI score and sex) to examine the relationship between functional network activation in the DMN and the BI measure. Sex and BDI scores were both covariates of no-interest. Significance was set at a threshold of p o 0.05 using FamilyWise Error correction (FWE).
1175
3. Results Data were collected from 30 youth: 15 with MDD; and 15 with BD, type I. Participants had no history of head injury or significant non-psychiatric medical illness. Details of the patient groups are illustrated in Table 1. There were no significant differences between the BD and MDD groups in age, illness duration or YMRS scores. As expected, medication differences existed between the groups, primarily in the use of antidepressants versus mood stabilizers (0 MDD, 12 BD). No patients were being treated with the older antipsychotics and a small number were receiving atypical antipsychotics. All BD patients had higher BI scores than MDD subjects, as would be expected, though there was a range in both groups, with greater variability in the BD patients (Table 1). Fig. 1 shows the whole-brain multiple regression analysis between resting-state DMN activation and the BI in all patients. Given the group differences in depression state we also conducted a multiple regression analysis between DMN network activation and individual scores on the BDI, while statistically controlling for the influence of BI score and sex. There were no voxels with either a significantly positive or negative correlation with the BDI score. Since there is a known relationship between activity in the putamen and movement, we conducted a post hoc examination for differences in movement between our groups during data acquisition. We examined six movement parameters calculated during data preprocessing (translation x, y, z; rotation pitch, roll, yaw), by conducting a t-test between groups for the total translation and total rotation of each subject. There were no significant group differences (translation: MDD mean 0.1 mm, SD 1.6; BD mean 0.6 mm, SD 1.4, rotation: MDD mean 0.0041, SD 0.04; BD mean 0.011, SD 0.02; p¼0.42). To help address the possibility that lithium use may have influenced these results, we performed post hoc analyses examining the relationship between functional activation in the DMN and lithium use in our BD patients. We found no significant differences in functional brain activation between those patients taking lithium (n ¼8) and those who were not (n¼ 7) (two-sample ttest), nor was there a relationship between lithium dose (mg/day) and functional activation in the DMN (multiple regression).
4. Discussion This study examined whole brain, resting-state DMN connectivity using fMRI correlated with BI in a group of MDD and BD youth. We found a significant positive correlation between DMN activation in the right putamen/claustrum extending into the insula and individual participant scores of trait bipolarity, and a significant negative correlation in left postcentral gyrus, extending into posterior cingulate cortex (PCC). Of note is the young age of the participants utilized in this study. The groups had an average age of approximately 20 years, with mean illness duration only 1.3 years. Higher resting-state functional activation correlated with BI localized to the putamen may be due to a relative decrease in functional connectivity in MDD in this region. A recent meta-analysis directly compared structural imaging results in MDD and BD and showed that MDD patients showing significantly reduced volumes in the putamen, as well as in the caudate, globus pallidus, and hippocampus (Kempton et al., 2011). Studies examining functional connectivity in MDD have shown altered connectivity between the PCC and bilateral caudate (Bluhm et al., 2009) and between the bilateral subcortical components of the putamen-thalamus (Marchand et al., 2012) suggesting these regions may be loci of primary pathology in MDD. Our findings are complementary to this extant literature. In addition to its function in limbic processes, the role of the putamen in movement is well known (Alexander et al., 1990). There
1176
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
Table 1 Demographic and clinical variables. BD n¼ 15 Mean (SD)
MDD n¼ 15 Mean (SD)
Statistics
p-value
Age at scan (years)
20.6 (2.6)
19.8 (2.6)
F(1,28) ¼ 0.9
0.41
Illness duration (months)
15.9 (14.4)
17.1 (17.7)
T(28) ¼0.22
0.47
Sex Male Female
10 5
4 11
χ2(1) ¼4.8
0.03
Medication Medication free Anticonvulsant mood stabilizers Lithium Atypical antipsychotics Antidepressants Benzodiazepines
3 4 8 5 3 1
20.0% 26.7% 53.3% 33.3% 20% 6.7%
2 0 0 1 13 1
13.3% 0% 0% 6.7% 86.7% 6.7%
Bipolarity index score
76.1 (15.0) Range 50–90
25.2 (6.2) Range 17–40
T(28) ¼−12.1
o0.0005
Beck depression inventory
9.9 (11.9)
28.53 (12.6)
T(28) ¼4.2
o0.0005
Young mania rating scale
2.0 (2.4)
2.5 (2.3)
T(28) ¼0.62
0.54
MDD¼ Major Depressive Disorder; BD ¼Bipolar Disorder, type I; SD ¼ standard deviation.
Fig. 1. Correlations between Bipolarity Index Score and DMN; significance threshold p o 0.05, cluster level FWE corrected. (A) Clusters with significant positive correlation with Bipolarity Index Score while controlling for the influence of BDI score and sex. The image was overlaid on a normalized T1-weighted anatomical image at the sagittal, transverse and coronal MNI planes of view indicated to clarify localization. The cluster illustrated fell within the right putamen/claustrum/insula with peak MNI coordinates of [36, 8, −8], t-score¼4.88 (281 of freedom and a z-score¼ 4.07. (B) Clusters with significant negative correlation with Bipolarity Index Score while controlling for the influence of BDI score and sex. The image overlaid on a normalized T1-weighted anatomical image at MNI planes of view indicated to clarify localization. The cluster illustrated fell within the left postcentral gyrus, extending into the posterior cingulate cortex (PCC), with peak MNI coordinates of [−26, −36, 48] t-score¼ 4.53, (281 of freedom), z-score¼ 3.86.
were no significant differences between our BD and MDD groups in movement during data acquisition, as shown above, nor in state mania. However, evidence does suggest that psychomotor activation, including thought acceleration, distractibility, hyperactivity and
restlessness may be a potent discriminator between BD and MDD (Cassano et al., 2012). Our finding of increased activation in the putamen during rest correlated with BI score, and thus trait bipolarity, may reflect a possible neural correlate of this clinical feature.
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
Importantly, work from a number of sources has shown specific abnormalities in the putamen associated with BD, including in young patients. Right putamen volume has been associated with impaired executive functioning in BD and schizophrenia relative to healthy controls (Hartberg et al., 2011). Increased activation in the putamen during facial emotional processing has been identified in BD across mood states (Hulvershorn et al., 2012), and adolescents with bipolar disorder have been shown to exhibit increased putamen volume (DelBello et al., 2004). In addition, limbic system hyperactivation, including in the putamen, has been implicated in bipolar youth during the processing of neutral facial expressions (Rich et al., 2006). The trait characterization of this correlation is also supported by work demonstrating that three separate groups of BD patients in differing mood states showed increased activation in the putamen during facial emotional processing (Hulvershorn et al., 2012). A recent meta-analysis of fMRI findings in BD also implicated abnormal putamen function, along with the inferior frontal cortex, medial temporal structures and basal ganglia (Chen et al., 2011), although it suggested an underactive putamen in BD compared with healthy controls. Our findings suggest that increased putamen connectivity with the DMN in the brain at rest may represent a marker for BD that could be helpful for distinguishing it from MDD. It should be noted that several studies have suggested that older but not atypical antipsychotics may be associated with basal ganglia enlargement (Chakos et al., 1995; Lang et al., 2004). None of our participants were taking older antipsychotics, and only a small number were taking atypical antipsychotics, which reduces the likelihood that our findings were related to this phenomenon. The negative correlation found here between activation and BI in the left postcentral gyrus, including the left PCC, is also consistent with previous work. Prior work has shown an association between postcentral gyrus hypo-activation in the restingstate and MDD (Kuhn and Gallinat, 2013; Liu et al., 2012). Still other researchers have identified reduced functional connectivity in PCC (Guo et al., 2013; Zhu et al., 2012) in MDD, consistent with our negative correlation between this area and BI in our mixed BD/ MDD patient group. Limitations of this study included group differences in statedepression and sex that needed to be statistically controlled. In addition, differences in medication use existed between the groups. Subgroup analysis suggested that lithium was not the primary reason for the positive correlation between BI and the putamen/claustrum cluster. We did not conduct a subgroup analysis for the MDD group with respect to the negative correlation and antidepressant use, however, since the specific antidepressant medications used differed widely among participants. Sample size in this study was not large. Although further research is needed, our findings support the possibility that resting-state, DMN functional connectivity in the putamen/claustrum and the postcentral gyrus/PCC may represent neurobiological nodes for characterizing the poles of a “spectrum” of affective disorders. Future studies will determine if functional connectivity of the DMN in these regions represent trait-markers for BD versus MDD.
Role of funding source Funding of this project was through grant support provided by the Lawson Health Research Institute with institutional support from the University of Western Ontario. No other source of monies, expertise or assistance for this research, nor the completion of this manuscript, was provided by any commercial interest.
Conflict of interest No author has any financial or other conflict of interest with any of the material presented in this manuscript. Dr. Osuch has, subsequent to this study, received an
1177
Independent Investigator Award from Pfizer, but none of the material in this manuscript relates to that funding.
Acknowledgments The authors would like to thank Drs. Sarah Armstrong, Katherine Macdonald, Volker Hocke and Daniel Hertzman for their assistance patient recruitment and assessment. Special thanks to John Butler for fMRI scanning support; Christeen Forster and Erin Ross for their clinical involvement and Andrew Wrath for his diligent assistance with data entry and organization for the project.
References Akiskal, H.S., Djenderedjian, A.M., Rosenthal, R.H., Khani, M.K., 1977. Cyclothymic disorder: validating criteria for inclusion in the bipolar affective group. American Journal of Psychiatry 134, 1227–1233. Alexander, G.E., Crutcher, M.D., DeLong, M.R., 1990. Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, “prefrontal” and “limbic” functions. Progress in Brain Research 85, 119–146. Angst, J., Sellaro, R., Stassen, H.H., Gamma, A., 2005. Diagnostic conversion from depression to bipolar disorders: results of a long-term prospective study of hospital admissions. Journal of Affective Disorders 84, 149–157. Beck, A.T., Steer, R.A., Brown, G.K., 1996. Manual for Beck Depression Inventory-II. Psychological Corporation, San Antonio. Berk, M., Hallam, K., Malhi, G.S., Henry, L., Hasty, M., Macneil, C., Yucel, M., Pantelis, C., Murphy, B., Vieta, E., Dodd, S., McGorry, P.D., 2010. Evidence and implications for early intervention in bipolar disorder. Journal of Mental Health 19, 113–126. Bluhm, R., Williamson, P., Lanius, R., Theberge, J., Densmore, M., Bartha, R., Neufeld, R., Osuch, E., 2009. Resting state default-mode network connectivity in early depression using a seed region-of-interest analysis: decreased connectivity with caudate nucleus. Psychiatry and Clinical Neurosciences 63, 754–761. Cassano, G.B., Rucci, P., Benvenuti, A., Miniati, M., Calugi, S., Maggi, L., Pini, S., Kupfer, D.J., Maj, M., Fagiolini, A., Frank, E., 2012. The role of psychomotor activation in discriminating unipolar from bipolar disorders: a classificationtree analysis. Journal of Clinical Psychiatry 73, 22–28. Chakos, M.H., Lieberman, J.A., Alvir, J., Bilder, R., Ashtari, M., 1995. Caudate nuclei volumes in schizophrenic patients treated with typical antipsychotics or clozapine. Lancet 345, 456–457. Chen, C.H., Suckling, J., Lennox, B.R., Ooi, C., Bullmore, E.T., 2011. A quantitative meta-analysis of fMRI studies in bipolar disorder. Bipolar Disorders 13, 1–15. DelBello, M.P., Zimmerman, M.E., Mills, N.P., Getz, G.E., Strakowski, S.M., 2004. Magnetic resonance imaging analysis of amygdala and other subcortical brain regions in adolescents with bipolar disorder. Bipolar Disorders 6, 43–52. Guo, W., Liu, F., Dai, Y., Jiang, M., Zhang, J., Yu, L., Long, L., Chen, H., Gao, Q., Xiao, C., 2013. Decreased interhemispheric resting-state functional connectivity in firstepisode, drug-naive major depressive disorder. Progress in Neuropsychopharmacolology and Biological Psychiatry 41, 24–29. Hartberg, C.B., Sundet, K., Rimol, L.M., Haukvik, U.K., Lange, E.H., Nesvag, R., Melle, I., Andreassen, O.A., Agartz, I., 2011. Subcortical brain volumes relate to neurocognition in schizophrenia and bipolar disorder and healthy controls. Progress in Neuro-Psychopharmacology and Biological Psychiatry 35, 1122–1130. Hulvershorn, L.A., Karne, H., Gunn, A.D., Hartwick, S.L., Wang, Y., Hummer, T.A., Anand, A., 2012. Neural activation during facial emotion processing in unmedicated bipolar depression, euthymia, and mania. Biological Psychiatry 71, 603–610. Kempton, M.J., Salvador, Z., Munafo, M.R., Geddes, J.R., Simmons, A., Frangou, S., Williams, S.C., 2011. Structural neuroimaging studies in major depressive disorder: meta-analysis and comparison with bipolar disorder. Archives of General Psychiatry 68, 675–690. Kuhn, S., Gallinat, J., 2013. Resting-state brain activity in schizophrenia and major depression: a quantitative meta-analysis. Schizophrenia Bulletin 39, 358–365. Lang, D.J., Kopala, L.C., Vandorpe, R.A., Ru, Q., Smith, G.N., Goghari, V.M., 2004. Reduced basal ganglia volumes after switching to olanzapine in chronically treated patients with schizophrenia. American Journal of Psychiatry 161, 1829–1836. Liu, F., Hu, M., Wang, S., Guo, W., Zhao, J., Li, J., Xun, G., Long, Z., Zhang, J., Wang, Y., Zeng, L., Gao, Q., Wooderson, S.C., Chen, J., Chen, H., 2012. Abnormal regional spontaneous neural activity in first-episode, treatment-naive patients with latelife depression: a resting-state fMRI study. Progress in Neuro-psychopharmacology and Biological Psychiatry 39, 326–331. Marchand, W.R., Lee, J.N., Suchy, Y., Johnson, S., Thatcher, J., Gale, P., 2012. Aberrant functional connectivity of cortico-basal ganglia circuits in major depression. Neuroscience Letters 514, 86–90. Rich, B.A., Vinton, D.T., Roberson-Nay, R., Hommer, R.E., Berghorst, L.H., McClure, E.B., Fromm, S.J., Pine, D.S., Leibenluft, E., 2006. Limbic hyperactivation during processing of neutral facial expressions in children with bipolar disorder. Proceedings of the National Academy of Sciences USA 103, 8900–8905. Sachs, G.S., 2004. Strategies for improving treatment of bipolar disorder: integration of measurement and management. Acta Psychiatrica Scandinavica Supplementum, 7–17. Shi, L., Thiebaud, P., McCombs, J.S., 2004. The impact of unrecognized bipolar disorders for patients treated for depression with antidepressants in the fee-
1178
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
for-services California Medicaid (Medi-Cal) program. Journal of Affective Disorders 82, 373–383. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. British Journal of Psychiatry 133, 429–435. Zhu, X., Wang, X., Xiao, J., Liao, J., Zhong, M., Wang, W., Yao, S., 2012. Evidence of a dissociation pattern in resting-state default mode network connectivity in first-
episode, treatment-naive major depression patients. Biological Psychiatry 71, 611–617. Zimmermann, P., Bruckl, 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 subthreshold bipolarity. Archives of General Psychiatry 66, 1341–1352.
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Preliminary communication
Correlation of brain default mode network activation with bipolarity index in youth with mood disorders Kristen A. Ford a, Jean Théberge b, Richard J. Neufeld a,c,d, Peter C. Williamson a,b, Elizabeth A. Osuch a,b,n a
University of Western Ontario, Schulich School of Medicine and Dentistry, Department of Psychiatry, London, Ontario, Canada University of Western Ontario, Department of Medical Biophysics, and Lawson Health Research Institute, London, Ontario, Canada c University of Western Ontario, Department of Psychology, London, Ontario, Canada d University of Western Ontario, Program in Neuroscience, London, Ontario, Canada b
art ic l e i nf o
a b s t r a c t
Article history: Received 27 March 2013 Received in revised form 22 May 2013 Accepted 31 May 2013 Available online 24 June 2013
Background: Major Depressive Disorder (MDD) and Bipolar Disorder (BD) can be difficult to differentiate, as both feature depressive episodes. Here we have utilized fMRI and a measure of trait bipolarity to examine resting-state functional connectivity of brain activation in the default mode network in youth with MDD and BD to isolate trait-specific patterns. Methods: We collected resting-state fMRI scans from thirty youth (15 MDD; 15 BD, Type 1). The Bipolarity Index (BI) was completed by each patient's treating psychiatrist. Independent components analysis was used to extract a default mode network component from each participant, and then multiple regression was used to identify correlations between bipolarity and network activation. Results: Activation in putamen/claustrum/insula correlated positively with BI; activation in the postcentral gyrus/posterior cingulate gyrus correlated negatively with BI. These correlations did not appear to be driven by movement in the scanner, state depression, gender or lithium use. Limitations: There were group differences in state depression and sex that needed to be statistically covaried; differences in medication use existed between the groups; sample size was not large. Conclusions: The identification of the putamen/claustrum in our positive correlation may indicate a potential trait marker for the psychomotor activation unique to bipolar mania. The negative correlation in the postcentral gyrus/posterior cingulate suggests that this functional inactivation is more specific to MDD and is consistent with previous research. Ultimately, this approach may help to develop techniques to minimize the current clinical dilemma by facilitating the classification between BD and MDD. & 2013 Elsevier B.V. All rights reserved.
Keywords: Bipolar Disorder Youth mental health Major Depressive Disorder Functional brain imaging Bipolar spectrum Default mode network
1. Introduction Major Depressive Disorder (MDD) and Bipolar Disorder (BD) both feature depressive episodes, with depression the most common presenting complaint for both in ambulatory settings. BD patients go an average of 6–10 years without the proper diagnosis, usually diagnosed with MDD (Berk et al., 2010). Correct identification is critical because antidepressant use in patients with BD can lead to worsening of symptoms (Berk et al., 2010; Shi et al., 2004). The question has been raised of whether affective illnesses lie along a continuum from MDD to BD. The construct of “bipolar spectrum disorders” was developed to account for patients without mania or hypomania who had mood cycling (Akiskal et al., 1977). Certainly evidence showing that half of patients originally meeting criteria for
n Correspondence to: Schulich School of Medicine and Dentistry, Department of Psychiatry, 860 Richmond Street, London, ON N6A 3H8; Canada. Tel.: +1 519 646 6000x65188; fax: +1 519 646 6211. E-mail address: [email protected] (E.A. Osuch).
0165-0327/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jad.2013.05.088
MDD converted to a form of BD over three decades of study (Angst et al., 2005) indicates the current challenges of long-term differentiation between these disorders. It has been suggested that, while there appears to be a form of depression that indicates a “pure” MDD (Zimmermann et al., 2009), the rest of the primary affective disorders may lie on a continuum with BD. The purpose of this study was to examine the extent to which a measure of “bipolarity” would correlate with functional network connectivity in youth recently diagnosed with either MDD or BD. We hypothesized that participants with these mood disorders would demonstrate differences in resting-state default mode network (DMN) connectivity that would be related to the construct known as bipolarity, which constitutes a continuous variable measurable across participant groups.
2. Methods Approval for the protocol was obtained from the Research Ethics Board at the University of Western Ontario, Schulich School
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
of Medicine and Dentistry. Participants were recruited from the First Episode Mood and Anxiety Program (FEMAP) in London, Ontario, Canada. After a complete description of the study to participants, written informed consent was obtained. For each participant, Bipolarity Index scores (BI) (Sachs, 2004) were completed by their treating psychiatrist. Three team psychiatrists completed the BI on 27 participants and 3 other psychiatrist did the remainder. The BI provides a score (between 0 and 100) using five “dimensions of bipolarity” including mood symptoms, age of onset of symptoms, course of illness, response to medications, and family history of mood and substance use. The instrument measures trait features, independent of mood state at any given time. Each participant also met diagnostic criteria for Bipolar Disorder, type I or Major Depressive Disorder using the Structured Clinical Interview for Diagnosis-IV. Medications were unchanged for three weeks prior to scanning. The Beck Depression Inventory (BDI)(Beck et al., 1996) and Young Mania Rating Scale (YMRS)(Young et al., 1978) were administered on scanning day. All magnetic resonance imaging (MRI) was performed using a 3.0 T MRI scanner (Siemens Verio, Erlangen, Germany) at the Lawson Health Research Institute, and a 32-channel phased-array head coil (Siemens). Whole-brain, T1-weighted, anatomical images with 1 mm isotropic spatial resolution were acquired as reference for spatial normalization. An anatomical image was used to select the orientation of functional MRI images 61 coronal to the AC–PC plane. Functional scans consisted of gradient-echo, echo-planar scans (TR ¼3 s, TE ¼30 ms, FOV ¼240 mm 240 mm, matrix ¼80 80, flip angle 901, 40 transverse slices, thickness¼ 3 mm) with no parallel acceleration, covering whole brain with an isotropic spatial resolution of 3 mm for a total time of 8 min, 9 s per run (161 brain volumes plus two discarded steadystate volumes). Participants were instructed to lie comfortably with their eyes closed and let their minds wander without falling asleep. All participants reported compliance with these instructions. fMRI data underwent standard pre-processing including motion correction, spatial normalization into the standard Montreal Neurological Institute (MNI) space and spatial smoothing with a 3D-Gaussian kernel (10 mm FWHM radius) in preparation for statistical analysis (SPM8). Whole-brain resting-state fMRI time courses of all participants were entered into a group spatial Independent Component Analyses (ICA). Using Group ICA of fMRI Toolbox (GIFT) software (http://mialab.mrn.org/software/gift/ index.html), this procedure decomposes the group data into maximally independent functional brain components by dimension estimation and data reduction with Principal Component Analysis followed by independent component estimation using the Infomax algorithm. This ICA analysis produced a series of 30 temporal components for each participant. Individual spatial maps were then constructed for each participant and converted to Z-scores. These represented the relative strengths of each component identified in the data of each participant. The DMN was identified as follows. Individual participant components were spatially sorted based on the strength of the correlation between each component and a standardized map of the DMN that was constructed to include Brodmann's areas 7, 10, 39, the precuneus, and the posterior cingulate. For each participant, the component with the highest correlation value to this map was selected as the DMN component-of-interest. A whole brain multiple regression analysis (SPM8) was conducted which incorporated three covariates (BDI score, BI score and sex) to examine the relationship between functional network activation in the DMN and the BI measure. Sex and BDI scores were both covariates of no-interest. Significance was set at a threshold of p o 0.05 using FamilyWise Error correction (FWE).
1175
3. Results Data were collected from 30 youth: 15 with MDD; and 15 with BD, type I. Participants had no history of head injury or significant non-psychiatric medical illness. Details of the patient groups are illustrated in Table 1. There were no significant differences between the BD and MDD groups in age, illness duration or YMRS scores. As expected, medication differences existed between the groups, primarily in the use of antidepressants versus mood stabilizers (0 MDD, 12 BD). No patients were being treated with the older antipsychotics and a small number were receiving atypical antipsychotics. All BD patients had higher BI scores than MDD subjects, as would be expected, though there was a range in both groups, with greater variability in the BD patients (Table 1). Fig. 1 shows the whole-brain multiple regression analysis between resting-state DMN activation and the BI in all patients. Given the group differences in depression state we also conducted a multiple regression analysis between DMN network activation and individual scores on the BDI, while statistically controlling for the influence of BI score and sex. There were no voxels with either a significantly positive or negative correlation with the BDI score. Since there is a known relationship between activity in the putamen and movement, we conducted a post hoc examination for differences in movement between our groups during data acquisition. We examined six movement parameters calculated during data preprocessing (translation x, y, z; rotation pitch, roll, yaw), by conducting a t-test between groups for the total translation and total rotation of each subject. There were no significant group differences (translation: MDD mean 0.1 mm, SD 1.6; BD mean 0.6 mm, SD 1.4, rotation: MDD mean 0.0041, SD 0.04; BD mean 0.011, SD 0.02; p¼0.42). To help address the possibility that lithium use may have influenced these results, we performed post hoc analyses examining the relationship between functional activation in the DMN and lithium use in our BD patients. We found no significant differences in functional brain activation between those patients taking lithium (n ¼8) and those who were not (n¼ 7) (two-sample ttest), nor was there a relationship between lithium dose (mg/day) and functional activation in the DMN (multiple regression).
4. Discussion This study examined whole brain, resting-state DMN connectivity using fMRI correlated with BI in a group of MDD and BD youth. We found a significant positive correlation between DMN activation in the right putamen/claustrum extending into the insula and individual participant scores of trait bipolarity, and a significant negative correlation in left postcentral gyrus, extending into posterior cingulate cortex (PCC). Of note is the young age of the participants utilized in this study. The groups had an average age of approximately 20 years, with mean illness duration only 1.3 years. Higher resting-state functional activation correlated with BI localized to the putamen may be due to a relative decrease in functional connectivity in MDD in this region. A recent meta-analysis directly compared structural imaging results in MDD and BD and showed that MDD patients showing significantly reduced volumes in the putamen, as well as in the caudate, globus pallidus, and hippocampus (Kempton et al., 2011). Studies examining functional connectivity in MDD have shown altered connectivity between the PCC and bilateral caudate (Bluhm et al., 2009) and between the bilateral subcortical components of the putamen-thalamus (Marchand et al., 2012) suggesting these regions may be loci of primary pathology in MDD. Our findings are complementary to this extant literature. In addition to its function in limbic processes, the role of the putamen in movement is well known (Alexander et al., 1990). There
1176
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
Table 1 Demographic and clinical variables. BD n¼ 15 Mean (SD)
MDD n¼ 15 Mean (SD)
Statistics
p-value
Age at scan (years)
20.6 (2.6)
19.8 (2.6)
F(1,28) ¼ 0.9
0.41
Illness duration (months)
15.9 (14.4)
17.1 (17.7)
T(28) ¼0.22
0.47
Sex Male Female
10 5
4 11
χ2(1) ¼4.8
0.03
Medication Medication free Anticonvulsant mood stabilizers Lithium Atypical antipsychotics Antidepressants Benzodiazepines
3 4 8 5 3 1
20.0% 26.7% 53.3% 33.3% 20% 6.7%
2 0 0 1 13 1
13.3% 0% 0% 6.7% 86.7% 6.7%
Bipolarity index score
76.1 (15.0) Range 50–90
25.2 (6.2) Range 17–40
T(28) ¼−12.1
o0.0005
Beck depression inventory
9.9 (11.9)
28.53 (12.6)
T(28) ¼4.2
o0.0005
Young mania rating scale
2.0 (2.4)
2.5 (2.3)
T(28) ¼0.62
0.54
MDD¼ Major Depressive Disorder; BD ¼Bipolar Disorder, type I; SD ¼ standard deviation.
Fig. 1. Correlations between Bipolarity Index Score and DMN; significance threshold p o 0.05, cluster level FWE corrected. (A) Clusters with significant positive correlation with Bipolarity Index Score while controlling for the influence of BDI score and sex. The image was overlaid on a normalized T1-weighted anatomical image at the sagittal, transverse and coronal MNI planes of view indicated to clarify localization. The cluster illustrated fell within the right putamen/claustrum/insula with peak MNI coordinates of [36, 8, −8], t-score¼4.88 (281 of freedom and a z-score¼ 4.07. (B) Clusters with significant negative correlation with Bipolarity Index Score while controlling for the influence of BDI score and sex. The image overlaid on a normalized T1-weighted anatomical image at MNI planes of view indicated to clarify localization. The cluster illustrated fell within the left postcentral gyrus, extending into the posterior cingulate cortex (PCC), with peak MNI coordinates of [−26, −36, 48] t-score¼ 4.53, (281 of freedom), z-score¼ 3.86.
were no significant differences between our BD and MDD groups in movement during data acquisition, as shown above, nor in state mania. However, evidence does suggest that psychomotor activation, including thought acceleration, distractibility, hyperactivity and
restlessness may be a potent discriminator between BD and MDD (Cassano et al., 2012). Our finding of increased activation in the putamen during rest correlated with BI score, and thus trait bipolarity, may reflect a possible neural correlate of this clinical feature.
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
Importantly, work from a number of sources has shown specific abnormalities in the putamen associated with BD, including in young patients. Right putamen volume has been associated with impaired executive functioning in BD and schizophrenia relative to healthy controls (Hartberg et al., 2011). Increased activation in the putamen during facial emotional processing has been identified in BD across mood states (Hulvershorn et al., 2012), and adolescents with bipolar disorder have been shown to exhibit increased putamen volume (DelBello et al., 2004). In addition, limbic system hyperactivation, including in the putamen, has been implicated in bipolar youth during the processing of neutral facial expressions (Rich et al., 2006). The trait characterization of this correlation is also supported by work demonstrating that three separate groups of BD patients in differing mood states showed increased activation in the putamen during facial emotional processing (Hulvershorn et al., 2012). A recent meta-analysis of fMRI findings in BD also implicated abnormal putamen function, along with the inferior frontal cortex, medial temporal structures and basal ganglia (Chen et al., 2011), although it suggested an underactive putamen in BD compared with healthy controls. Our findings suggest that increased putamen connectivity with the DMN in the brain at rest may represent a marker for BD that could be helpful for distinguishing it from MDD. It should be noted that several studies have suggested that older but not atypical antipsychotics may be associated with basal ganglia enlargement (Chakos et al., 1995; Lang et al., 2004). None of our participants were taking older antipsychotics, and only a small number were taking atypical antipsychotics, which reduces the likelihood that our findings were related to this phenomenon. The negative correlation found here between activation and BI in the left postcentral gyrus, including the left PCC, is also consistent with previous work. Prior work has shown an association between postcentral gyrus hypo-activation in the restingstate and MDD (Kuhn and Gallinat, 2013; Liu et al., 2012). Still other researchers have identified reduced functional connectivity in PCC (Guo et al., 2013; Zhu et al., 2012) in MDD, consistent with our negative correlation between this area and BI in our mixed BD/ MDD patient group. Limitations of this study included group differences in statedepression and sex that needed to be statistically controlled. In addition, differences in medication use existed between the groups. Subgroup analysis suggested that lithium was not the primary reason for the positive correlation between BI and the putamen/claustrum cluster. We did not conduct a subgroup analysis for the MDD group with respect to the negative correlation and antidepressant use, however, since the specific antidepressant medications used differed widely among participants. Sample size in this study was not large. Although further research is needed, our findings support the possibility that resting-state, DMN functional connectivity in the putamen/claustrum and the postcentral gyrus/PCC may represent neurobiological nodes for characterizing the poles of a “spectrum” of affective disorders. Future studies will determine if functional connectivity of the DMN in these regions represent trait-markers for BD versus MDD.
Role of funding source Funding of this project was through grant support provided by the Lawson Health Research Institute with institutional support from the University of Western Ontario. No other source of monies, expertise or assistance for this research, nor the completion of this manuscript, was provided by any commercial interest.
Conflict of interest No author has any financial or other conflict of interest with any of the material presented in this manuscript. Dr. Osuch has, subsequent to this study, received an
1177
Independent Investigator Award from Pfizer, but none of the material in this manuscript relates to that funding.
Acknowledgments The authors would like to thank Drs. Sarah Armstrong, Katherine Macdonald, Volker Hocke and Daniel Hertzman for their assistance patient recruitment and assessment. Special thanks to John Butler for fMRI scanning support; Christeen Forster and Erin Ross for their clinical involvement and Andrew Wrath for his diligent assistance with data entry and organization for the project.
References Akiskal, H.S., Djenderedjian, A.M., Rosenthal, R.H., Khani, M.K., 1977. Cyclothymic disorder: validating criteria for inclusion in the bipolar affective group. American Journal of Psychiatry 134, 1227–1233. Alexander, G.E., Crutcher, M.D., DeLong, M.R., 1990. Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, “prefrontal” and “limbic” functions. Progress in Brain Research 85, 119–146. Angst, J., Sellaro, R., Stassen, H.H., Gamma, A., 2005. Diagnostic conversion from depression to bipolar disorders: results of a long-term prospective study of hospital admissions. Journal of Affective Disorders 84, 149–157. Beck, A.T., Steer, R.A., Brown, G.K., 1996. Manual for Beck Depression Inventory-II. Psychological Corporation, San Antonio. Berk, M., Hallam, K., Malhi, G.S., Henry, L., Hasty, M., Macneil, C., Yucel, M., Pantelis, C., Murphy, B., Vieta, E., Dodd, S., McGorry, P.D., 2010. Evidence and implications for early intervention in bipolar disorder. Journal of Mental Health 19, 113–126. Bluhm, R., Williamson, P., Lanius, R., Theberge, J., Densmore, M., Bartha, R., Neufeld, R., Osuch, E., 2009. Resting state default-mode network connectivity in early depression using a seed region-of-interest analysis: decreased connectivity with caudate nucleus. Psychiatry and Clinical Neurosciences 63, 754–761. Cassano, G.B., Rucci, P., Benvenuti, A., Miniati, M., Calugi, S., Maggi, L., Pini, S., Kupfer, D.J., Maj, M., Fagiolini, A., Frank, E., 2012. The role of psychomotor activation in discriminating unipolar from bipolar disorders: a classificationtree analysis. Journal of Clinical Psychiatry 73, 22–28. Chakos, M.H., Lieberman, J.A., Alvir, J., Bilder, R., Ashtari, M., 1995. Caudate nuclei volumes in schizophrenic patients treated with typical antipsychotics or clozapine. Lancet 345, 456–457. Chen, C.H., Suckling, J., Lennox, B.R., Ooi, C., Bullmore, E.T., 2011. A quantitative meta-analysis of fMRI studies in bipolar disorder. Bipolar Disorders 13, 1–15. DelBello, M.P., Zimmerman, M.E., Mills, N.P., Getz, G.E., Strakowski, S.M., 2004. Magnetic resonance imaging analysis of amygdala and other subcortical brain regions in adolescents with bipolar disorder. Bipolar Disorders 6, 43–52. Guo, W., Liu, F., Dai, Y., Jiang, M., Zhang, J., Yu, L., Long, L., Chen, H., Gao, Q., Xiao, C., 2013. Decreased interhemispheric resting-state functional connectivity in firstepisode, drug-naive major depressive disorder. Progress in Neuropsychopharmacolology and Biological Psychiatry 41, 24–29. Hartberg, C.B., Sundet, K., Rimol, L.M., Haukvik, U.K., Lange, E.H., Nesvag, R., Melle, I., Andreassen, O.A., Agartz, I., 2011. Subcortical brain volumes relate to neurocognition in schizophrenia and bipolar disorder and healthy controls. Progress in Neuro-Psychopharmacology and Biological Psychiatry 35, 1122–1130. Hulvershorn, L.A., Karne, H., Gunn, A.D., Hartwick, S.L., Wang, Y., Hummer, T.A., Anand, A., 2012. Neural activation during facial emotion processing in unmedicated bipolar depression, euthymia, and mania. Biological Psychiatry 71, 603–610. Kempton, M.J., Salvador, Z., Munafo, M.R., Geddes, J.R., Simmons, A., Frangou, S., Williams, S.C., 2011. Structural neuroimaging studies in major depressive disorder: meta-analysis and comparison with bipolar disorder. Archives of General Psychiatry 68, 675–690. Kuhn, S., Gallinat, J., 2013. Resting-state brain activity in schizophrenia and major depression: a quantitative meta-analysis. Schizophrenia Bulletin 39, 358–365. Lang, D.J., Kopala, L.C., Vandorpe, R.A., Ru, Q., Smith, G.N., Goghari, V.M., 2004. Reduced basal ganglia volumes after switching to olanzapine in chronically treated patients with schizophrenia. American Journal of Psychiatry 161, 1829–1836. Liu, F., Hu, M., Wang, S., Guo, W., Zhao, J., Li, J., Xun, G., Long, Z., Zhang, J., Wang, Y., Zeng, L., Gao, Q., Wooderson, S.C., Chen, J., Chen, H., 2012. Abnormal regional spontaneous neural activity in first-episode, treatment-naive patients with latelife depression: a resting-state fMRI study. Progress in Neuro-psychopharmacology and Biological Psychiatry 39, 326–331. Marchand, W.R., Lee, J.N., Suchy, Y., Johnson, S., Thatcher, J., Gale, P., 2012. Aberrant functional connectivity of cortico-basal ganglia circuits in major depression. Neuroscience Letters 514, 86–90. Rich, B.A., Vinton, D.T., Roberson-Nay, R., Hommer, R.E., Berghorst, L.H., McClure, E.B., Fromm, S.J., Pine, D.S., Leibenluft, E., 2006. Limbic hyperactivation during processing of neutral facial expressions in children with bipolar disorder. Proceedings of the National Academy of Sciences USA 103, 8900–8905. Sachs, G.S., 2004. Strategies for improving treatment of bipolar disorder: integration of measurement and management. Acta Psychiatrica Scandinavica Supplementum, 7–17. Shi, L., Thiebaud, P., McCombs, J.S., 2004. The impact of unrecognized bipolar disorders for patients treated for depression with antidepressants in the fee-
1178
K.A. Ford et al. / Journal of Affective Disorders 150 (2013) 1174–1178
for-services California Medicaid (Medi-Cal) program. Journal of Affective Disorders 82, 373–383. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. British Journal of Psychiatry 133, 429–435. Zhu, X., Wang, X., Xiao, J., Liao, J., Zhong, M., Wang, W., Yao, S., 2012. Evidence of a dissociation pattern in resting-state default mode network connectivity in first-
episode, treatment-naive major depression patients. Biological Psychiatry 71, 611–617. Zimmermann, P., Bruckl, 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 subthreshold bipolarity. Archives of General Psychiatry 66, 1341–1352.