- Email: [email protected]
Pain and emotion in the insular cortex: evidence for functional reorganization in major depression
Neuroscience Letters 520 (2012) 204–209
Contents lists available at SciVerse ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Review
Pain and emotion in the insular cortex: evidence for functional reorganization in major depression Isabella Mutschler a,b,∗,1 , Tonio Ball c,d,1 , Johanna Wankerl b , Irina A. Strigo a,e,f a
Department of Psychiatry, University of California San Diego, La Jolla, CA, USA Department of Psychology, Clinical Psychology and Epidemiology, University of Basel, Switzerland Bernstein Center Freiburg, University of Freiburg, Germany d Epilepsy-Center, University Medical Center Freiburg, Germany e Research Service, VA San Diego Healthcare System, San Diego, CA, USA f VA Center of Excellence for Stress and Mental Health (CESAMH), San Diego, CA, USA b c
a r t i c l e
i n f o
Article history: Received 27 December 2011 Received in revised form 27 March 2012 Accepted 30 March 2012 Keywords: Pain Emotion Depression Emotional allodynia Functional reorganization Nociception Psychophathology Meta-analysis Insula Interoception
a b s t r a c t Major Depressive Disorder (MDD) is among the top causes of disability worldwide and many patients with depression experience pain symptoms. Little is known regarding what makes depressed persons feel like they are in pain. An increasing number of neuroimaging studies show that both physical pain and depression involve the insular cortex. The present study aimed to investigate whether emotional processing in MDD patients is topologically shifted towards the insular area(s) involved in pain processing in healthy individuals. To achieve this aim, we investigated the functional organization of the insula by conducting meta-analyses of previously published neuroimaging studies on: (1) emotion in patients with MDD, (2) emotion in healthy subjects, and (3) physical pain in healthy subjects. Our results show that the dorsal part of the insula is reproducibly activated during experimental pain in healthy individuals, with multiple separate pain-related areas aligned along its dorsal border. Regions with maximal pain-related activation likelihood estimate (ALE) were located in the posterior (left) and dorsal mid-anterior insula (left and right). Furthermore, emotion-related peaks in healthy subjects were found both in its ventral (as shown in a previous meta-analysis) and dorsal anterior part. Importantly, emotion-related peaks in depressed patients were shifted to the dorsal anterior insula, where regions related to physical pain in healthy subjects are located. This shift was reflected in the observation that median z-coordinates of emotion-related responses in the left hemisphere were significantly larger in depressed patients than in healthy controls. This shift of emotion-related responses to the dorsal insula, i.e., where pain-processing takes place in healthy subjects, may play a role in “emotional allodynia” – a notion that individuals with MDD experience pain in response to stimuli that are normally not painful. © 2012 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3. 4. 5.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Search criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author at: Department of Psychiatry, University of California San Diego, La Jolla, CA, USA. Tel.: +1 858 230 4323; fax: +1 858 642 1601. E-mail address: [email protected] (I. Mutschler). 1 These authors contributed equally. 0304-3940/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neulet.2012.03.095
205 205 205 205 205 205 206 208 208 208
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
205
1. Introduction
2.2. Inclusion criteria
Major depressive disorder (MDD) is one of the most common psychiatric disorders – with a lifetime prevalence for MDD as high as 15% worldwide [16]. It is defined as a period of at least two weeks of sustained depressed mood and/or anhedonia [16]. MDD profoundly impacts the individual’s life and health in general and has been associated with a number of adverse health outcomes including a heightened subjective experience of pain [31]. When William Styron described his depression, he said: “Mysteriously and in ways that are totally remote from natural experience, the gray drizzle of horror induced by depression takes on the quality of physical pain.” Over 75% of patients with depression experience pain symptoms [19]. Conversely, 30–60% of chronic pain patients report significant depressive symptoms [3]. The presence of comorbid pain contributes significantly to unfavorable outcomes and increased cost of treatment in MDD [13]. Little is known, however, about the neuronal mechanisms that make depressed individuals feel like they are in pain. A large number of neuroimaging studies have investigated functional brain changes in depression and physical pain. Experience of pain has been associated with activation in the insula and the anterior cingulate cortex, as well as somatosensory areas and prefrontal regions – a network often referred to as “the pain matrix” [33]. A recent meta-analysis shows that responses to emotional stimuli in depression involve the insula, prefrontal cortices, the hippocampus, the amygdala, the anterior cingulate cortex, the cerebellum, the thalamus, the caudate nucleus, the putamen and the globus pallidus [9]. Together, these findings indicate that both physical pain and depression involve the insula. Recently, it has been shown that altered activity in the insula is associated with abnormal interoception [36] and pain processing in MDD [4,14,15,21,30,31]. As the insula is strongly involved both in pain and emotional processing, in the present study, we have investigated whether there are alterations in insular activation that may help explaining “emotional allodynia”– that is why somatosensory stimuli that are typically perceived as non-painful in healthy subjects can cause emotional pain in depressed patients [32]. As a possible neural correlate to this clinical observation, we put forward the hypothesis that emotional processing in depressed patients is topologically shifted toward the insular area (s) associated with physical pain in healthy individuals. Hence, this study had the aim to investigate the functional organization of the insula by conducting activation likelihood estimate (ALE) meta-analyses of previously published neuroimaging studies on: (1) emotion in individuals with MDD, (2) emotion, and (3) physical pain in healthy subjects. As has been recently shown, meta-analyses are useful tools for detecting brain responses that are reproducibly observed across functional imaging studies and for investigating the functional organization of the human insula [25,27]. In addition, for data interpretation, an ALE on studies investigating sensorimotor tasks was carried out (for more details see reference [27]). Furthermore, we provide a brief overview of the few neuroimaging studies that examined experimental pain processing in individuals with MDD.
We included studies that: (1) investigated emotional processing or experimentally induced physical pain in healthy individuals and emotional processing in patients with MDD using functional magnetic resonance imaging (fMRI) or positron emission tomography (PET); (2) reported insular peaks that lay within 5 mm of the insular border (for more details see below); (3) provided Talairach or Montreal Neurological Institute (MNI) coordinates; (4) included hand movement (e.g., button press) only if two experimental conditions assumed to have equal movement-related activity were contrasted; this was selected to rule out movement-related effects, which are known to activate a region in the anterior insular cortex (see Fig. 1 and references [25,27]); (5) examined healthy individuals that did not suffer from any neurological or psychiatric disorder; (6) examined depressed patients meeting the criteria for MDD based on the Diagnostic and Statistical Manual, fourth edition (DSM-IV [1]) or the International Classification of Diseases, tenth revision (ICD-10 [2]).
2. Methods 2.1. Search criteria A systematic literature search in large databases (Medline, PsychInfo) for neuroimaging studies published in English prior to December 2011 was conducted. The search keywords were “depressive”; “depression”; “MDD”; “pain”; “noxious stimuli”; “emotion”; “brain imaging”; “fMRI”; “PET”. In addition; we examined the references of the selected papers.
3. Data analysis Descriptive information was extracted from each article. Talairach coordinates were translated to match the MNI space using the tal2mni Matlab script from http://eeg.sourceforge.net/ doc m2html/bioelectromagnetism/tal2mni.html. An ‘activation likelihood estimate’ (ALE), given by the union of the probabilities associated with the different foci, was calculated for an area comprising the whole Y and Z extent of the insular cortex. ALE was calculated across studies such that the summed ALE of all peaks of each study was normalized to unity to ensure that studies reporting large numbers of peaks cannot disproportionally dominate the resulting ALE map. The average extent of the insular cortex was determined from the T1 multi-subject template provided with SPM5. The anatomical boundaries of the insula are described in the study by Makris et al. [22]. The spatial union of all ALE values could be portrayed as an ALE-value map describing the reproducibility of an effect within different insular subregions. Statistical significance was assessed using a (non-parametric) single-threshold permutation test as previously described [18,34]. ALE scores inferred from the distribution of activation foci from the included studies of one functional category were compared against a single critical ALE score derived from an empirical null distribution. This null distribution resulted from 10,000 permutations of an equal number of foci that were distributed randomly throughout the insular cortex. The ALE maps that resulted from the separate analyses of different functional categories are presented. The following ALE results are reported at p < 0.05 (FDR corrected): fMRI studies on pain in healthy individuals, and studies on sensorimotor tasks (Fig. 1, red and turquoise). The remaining ALE results are reported at p < 0.001 (uncorrected), because there were no significant ALE values at the FDR-corrected threshold. Data were processed using SPM5/8 alongside with in-house developed MATLAB (Version 7.0.4, MathWorks) scripts. To evaluate whether response peaks in depressed patients were shifted dorsally (i.e., where reproducible pain-related responses were found, see below), we applied a Wilcoxon rank sum test to the z-coordinates of the reported peaks, both for those in the right and left hemisphere. This test was also applied to the y-coordinates. 4. Results Eleven fMRI studies on emotion and MDD (comprising 209 patients and 20 peaks), 44 studies on emotion in healthy subjects (comprising 756 subjects and 89 peaks), and 57 studies
206
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
Fig. 1. Shows activation likelihood estimate (ALE) findings for the left (a) and (b) for the right insula and their MNI z coordinates (on the y-axis) and y coordinates (on the x-axis) related to physical pain in healthy individuals (using fMRI = red, or PET = magenta), emotional processing in healthy subjects (yellow), emotional processing in depressed individuals (green), and sensorimotor tasks (turquoise). Studies on emotion in healthy individuals entering this ALE analysis did not include insula-coordinates which were related to the measurement of peripheral physiological changes. In a recent meta-analysis responses related to peripheral physiological changes resulting from emotional experiences were located in the ventral anterior insular cortex [27]. The solid gray line indicates the mean outline of the left sagittal insula and the dashed gray line the central sulcus of the insula dividing the insula in an anterior and posterior part (see methods for more details). In the left insular cortex, pain-related maximal ALE were found in the posterior and dorsal mid-anterior insula, whereas in the right insula the maximal ALE was located in the dorsal mid-anterior insula. Insula effects related to emotional processing in depressed individuals overlapped with the pain (left) and also sensorimotor areas (right) found in healthy subjects. Hand-movement task had their ALE maxima in the posterior part of the anterior insular cortex. This region has also been implicated in sense of agency of hand movement [10]. The reported ‘sense of agency’ related response peaks overlapped with the ALE maximum related to sensorimotor tasks (for more details see reference [27]).
on physical pain in healthy individuals (comprising 690 subjects and 175 peaks) fulfilled the inclusion criteria. Studies on pain predominantly used thermal heat stimuli (n = 29). Thirteen studies used positron emission tomography (PET) and 44 studies used fMRI. Studies on emotion used PET (N = 13) or fMRI (n = 31). Seven out of 11 studies on emotion in MDD investigated unmedicated patients. Eight studies on emotion in MDD specified that the patients and the healthy controls where matched. See Supplement for detailed description of the articles. The results of the ALE meta-analyses are shown in Fig. 1. We found that pain-related maximal ALE in healthy subjects was located in the posterior and dorsal mid-anterior insula on the left and in the dorsal mid-anterior insula on the right (Fig. 1). We further found that emotion-related maximal ALE in healthy individuals – excluding activation peaks in the insula related to the measurement of peripheral physiological changes – was located in the dorsal anterior insula. Conversely, emotion-related maximal ALE in individuals with MDD was located more dorsal and overlapped with the insula region related more to pain and sensorimotor processing in healthy subjects (Fig. 1). In other words, there was a shift of emotion-related responses in MDD towards the dorsal insula, a region where pain-processing takes place in healthy individuals. Furthermore, median z-coordinates of emotion-related responses were significantly larger in the depressed patients than in the healthy controls, in the left (p = 0.009), but not significantly in the right (p = 0.505, Fig. 2) hemisphere. The observed difference in the left insula prevailed independently of the coordinate system used in the studies (and therefore, independent of possible errors in reporting the coordinate system). Median y-coordinates of emotion-related responses were not significantly different between depressed patients and healthy controls: p = 0.547 (left) and p = 0.401 (right). We found that pain-related coordinates in healthy individuals from PET studies were located more posterior within the insular region than those from the fMRI studies. An ALE analysis on pain processing in MDD patients was not possible due to limited studies available to date.
Fig. 2. Boxplots showing the distribution of z-coordinates of emotion-related responses in depressed patients and in healthy subjects. Emotion-related peaks in depressed patients were shifted to the dorsal insula: all peaks in depressed patients were found in the dorsal part of the insula (z > 0) while, in healthy subjects, peaks were found both in the dorsal and in the ventral parts of the insula. Median zcoordinates of emotion-related responses were significantly larger in the depressed patients than in the healthy controls in the left hemisphere (p = 0.009), but not in the right (p = 0.505). This shift of emotion-related responses to the dorsal insula in MDD, i.e., where pain-processing takes place, may play a role in “emotional allodynia” – a notion that individuals with MDD hurt in response to stimuli that are not painful.
5. Discussion The main goal of this study was to investigate the hypothesis that emotional processing within the insula in depressed patients is topologically shifted toward the insular area(s) associated with physical pain in healthy individuals. Our results show that, compared to healthy subjects, emotional processing within the insula in MDD is shifted to the dorsal anterior insula, to regions including those with the maximal ALE related to pain processing in
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
healthy subjects. In support of this we found that the median zcoordinates of emotion-related responses in depressed patients compared to healthy controls were significantly larger on the left and non-significantly on the right. Taken together, the observed shift of emotion-related responses in MDD to the dorsal insula, i.e., where pain processing takes place in healthy subjects, may play a role in “emotional allodynia” – a notion that individuals with MDD hurt in response to stimuli that are not painful [32]. Our results show that many insular response peaks related to emotional processing in MDD overlapped with the maximal ALE related to sensorimotor tasks. Sensorimotor tasks reproducibly activate the posterior part of the anterior insular cortex, just anterior to the central sulcus of the insula. We previously related the reproducible movement-related activity in the anterior insula to the existence of the functional insular motor area, i.e., an area with neural processing directly involved in efferent motor control [27]. This interpretation is supported by a recent study that investigated connectivity patterns of the insula and found that the insular region involved in sensorimotor tasks showed functional connections with the premotor cortex (BA 6) that is involved in motor control [5]. Alternatively, it has been suggested that the insula contains a somatotopic representation of the subjective feelings of one’s current movements as part of a representation of all feelings from the body [7]. The anterior insula has been reported to be involved in the ‘sense of agency’ of hand movement, that is, the experience of oneself being the cause of an action [10]. The reported response peaks related to the ‘sense of agency’ overlapped with the ALE maximum associated with sensorimotor tasks [27]. We found that emotion-related peaks in healthy subjects were located in the dorsal anterior insula. In a recent meta-analysis we examined insula response specifically related to peripheral physiological changes resulting from emotional experiences and found that this activity was represented in the ventral anterior insular cortex [27]. This region was also the most likely site of insular co-activation with the amygdala. In the current study, insula-coordinates associated with emotion-related peripheral physiological changes were excluded. Therefore, our findings suggest that distinct functional insula regions may be involved in different aspects of emotional processing, whereby peripheral physiological aspects of emotional processing are mapped to the ventral anterior regions, while more cognitive–evaluative aspects are mapped to the dorsal-anterior region. This interpretation is supported by the fact that in our study-sample subjects evaluated the emotional content of the presented stimuli. This conclusion is also in line with the observed functional connectivity between the dorsal anterior insula and the dorsal anterior cingulate cortex, a network that plays a crucial role in cognitive decision-making [8]. This is also consistent with earlier findings in monkeys show that the insula is interconnected with the cingulate cortex [23]. We found that pain-related peaks in healthy individuals were located in the left posterior and dorsal mid-anterior insula, whereas in the right insula pain-related representation was located in the dorsal mid-anterior insula. Similarly, a study by Schweinhardt et al. found that experimental pain in healthy volunteers was predominantly represented in the dorsal anterior and in the posterior insula [29]. The authors suggested, that posterior insula activation in experimental pain reflect basic sensory aspects of nociceptive input, which is consistent with the animal work and the proposed role of this region as “interoceptive sensory cortex” [37]. Furthermore, we found that within the right dorsal mid-anterior insula pain, emotion, and sensorimotor related responses overlap. Our findings are thus in direct agreement with the notion that this insula region plays an integrative role by
207
bringing together information for emotional–cognitive evaluation of a noxious stimuli and the associated motor reactions. Anatomical studies in monkeys show that the anterior insula – with its reciprocal connections to the cingulate cortex, the primary somatosensory cortex, and the prefrontal cortex – is in an ideal position to associate sensory events with emotional–cognitive and motor responses during the experience of pain [11,12,23,24]. Recently, Peltz et al. investigated the functional connectivity of insular cortex subregions during noxious and innocuous thermal stimulation [28]. The authors found that the anterior part of the insula was more functionally connected to areas known for affective and cognitive processing (including the anterior cingulate cortex and the prefrontal cortex), whereas the posterior part was more connected with areas known for sensory-discriminative processing of noxious and somatosensory stimuli (including the primary somatosensory cortex). Together, the findings of the present study indicate multiple representations of pain in the dorsal insula, arranged along the anterior–posterior axis along the dorsal border of the insula. Furthermore, our ALE findings are also in agreement with the idea of a posterior to anterior functional gradient, assuming that sensory information about the body’s physiological state is mapped in the posterior insula and re-represented in the anterior insula where it becomes consciously accessible, enabling a general subjective affective experience [7]. Few brain imaging studies to date have examined pain processing in individuals with MDD [4,14,15,21,30,31], therefore ALE analysis could not be performed. However, despite differences in experimental design and the imaging technique used, all six studies found increased pain-related processing within the insula in MDD further conforming insula dysregulation in this disorder. We briefly describe them: Graff-Guerrero et al. [15] conducted a Single Photon Emission Computed Tomography (SPECT) study in 24 unmedicated individuals with MDD who were all free of pain symptoms. They compared brain activation during application of painful pressure stimuli to the right distal phalanx of the fifth finger versus sham stimulation and found increased activation within left anterior insula and within more posterior aspects of the bilateral insula. We have shown that in healthy individuals pain-related peaks from PET studies are located more posterior within the insular cortex than those from the fMRI studies, which is consistent with the reported regionally specific sensitivity differences between fMRI and PET described previously [35]. Therefore, it is difficult to infer how insula activation from the Graff-Guerrero et al. [15] study maps to emotional processing insular region in MDD found here. In an fMRI study by Bar et al. [4] brain activation during experimental heat pain processing (right volar wrist) was compared between 13 women with MDD who were all unmedicated and free of pain symptoms and 13 healthy control women. The authors found increased brain response in MDD within dorsal anterior insula on the right and more medial ventral insula on the left. Strigo et al. [31] conducted an fMRI study comparing heat pain anticipation and processing (left volar forearm) in 15 subjects with MDD who were all unmedicated and free of pain symptoms to 15 matched healthy comparison subjects who never had MDD. The authors found increased activation during pain anticipation within bilateral dorsal anterior insula in MDD. Similarly, Strigo et al. [30] found right anterior insula dysfunction in MDD. Lopez-Sola et al. [21] conducted an fMRI study comparing heat pain processing (right volar forearm) in 13 unmedicated subjects with MDD who experienced low levels of baseline pain to 20 matched healthy subjects. The authors found significantly increased activation to pain in the MDD group within more posterior aspects of the bilateral insula. Finally, Giesecke et al. [14] examined pressure pain processing (left thumb) between 53 patients with fibromyalgia with significant depressive symptoms and 42 healthy control subjects. They found
208
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
that increased activation within dorsal anterior and more posterior insula on the right was uniquely associated with depression in their sample. Together, these studies clearly show that insula plays an important role during pain processing in individuals with MDD. It would be important to examine how pain-related activation in MDD maps to functional organization within insula cortex reported here when more studies become available. Several important limitations should be noted. Four studies on emotion and MDD investigated patients taking antidepressants, which are known to influence insula function [9]. Thus, our findings should be replicated when more studies on emotional processing in unmedicated MDD subjects become available. Furthermore, future studies should directly compare different ALE maps and examine whether there are significant laterality effects; we are beginning to address these questions [26]. Specifically, the dorsal shift in the emotion-related activation peak in MDD was significant on the left but not on the right side in our study. Interestingly, although non-significant on both sides, a visually apparent trend in the antero-posterior direction was noted in MDD on the right. It will be important to examine whether functional reorganization within the insula in MDD is lateralized, consistent with emotional asymmetry theories proposed by others [6]. Another relevant topic for further investigations would be the examination of the relationship between functionally defined areas – as investigated in the present study – and macro-anatomical landmarks of the human insula. To our knowledge no study has been published to date on the cytoarchitectonic arrangement of the human anterior insula. Finally, an ALE analysis on experimentally induced pain in MDD was not possible because to our knowledge, only six neuroimaging studies have been published to date (see above). To summarize, although pain is not a diagnostic symptom of MDD according to DSM-IV [1], it should be taken into account because over 75% of patients with depression experience pain symptoms [19]. Pain in depression has a strong negative impact on the clinical course, treatment response [17], and remission [20]. By applying a coordinate-based meta-analytical approach we found that emotion-related peaks in depressed individuals were shifted to the dorsal insula, towards the activation peaks associated with pain processing in healthy subjects. To our knowledge this is the first study which finds evidence for functional reorganization of the insula in depressed individuals, which may contribute to the neurobiological understanding of “emotional allodynia” – a notion when typically non-painful stimuli can cause emotional pain in depressed patients [32]. Does emotional pain physically hurt if one is depressed? The overlap between emotional and pain-processing circuitry found in the current study in people with MDD supports this notion. Future studies should examine whether such functional shift within the insula in MDD can be experimentally replicated and whether a similar effect might exist in other brain regions. It will be important to investigate whether this shift in emotional processing circuitry towards pain-processing circuitry within the insula is a vulnerability factor for the development of psychiatric and/or psychosomatic disorders and whether it remains after successful treatment. Conversely, it would be interesting to see whether segregation of pain-related and emotion-related activation peaks within the insula may indicate resilience for the development of psychopathology.
Acknowledgements We would like to thank Dr. Bud Craig for his advice on insula anatomy. We gratefully acknowledge the financial support by the Swiss National Science Foundation (SNF fellowship PA00P1 136408 to IM) and the National Institute of Mental Health (MH80003 to IAS).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.neulet.2012.03.095.
References [1] Diagnostic and statistical manual of mental disorders, 4th ed., American Psychiatric Association, Washington, 1994. [2] International Classification of Diseases (ICD), 10th revision, World Health Organization, 1996. [3] M.J. Bair, R.L. Robinson, W. Katon, K. Kroenke, Depression and pain comorbidity: a literature review, Archives of Internal Medicine 163 (2003) 2433–2445. [4] K.J. Bar, G. Wagner, M. Koschke, S. Boettger, M.K. Boettger, R. Schlosser, H. Sauer, Increased prefrontal activation during pain perception in major depression, Biological Psychiatry 62 (2007) 1281–1287. [5] L. Cerliani, R.M. Thomas, S. Jbabdi, J.C. Siero, L. Nanetti, A. Crippa, V. Gazzola, H. D’Arceuil, C. Keysers, Probabilistic tractography recovers a rostrocaudal trajectory of connectivity variability in the human insular cortex, Human Brain Mapping (2011) [Epub ahead of print]. [6] A.D. Craig, Forebrain emotional asymmetry: a neuroanatomical basis? Trends in Cognitive Science 9 (2005) 566–571. [7] A.D. Craig, How do you feel – now? The anterior insula and human awareness, Nature Reviews. Neuroscience 10 (2009) 59–70. [8] B. Deen, N.B. Pitskel, K.A. Pelphrey, Three systems of insular functional connectivity identified with cluster analysis, Cerebral Cortex 21 (2011) 1498–1506. [9] P. Delaveau, M. Jabourian, C. Lemogne, S. Guionnet, L. Bergouignan, P. Fossati, Brain effects of antidepressants in major depression: a meta-analysis of emotional processing studies, Journal of Affective Disorders 130 (2011) 66–74. [10] C. Farrer, C.D. Frith, Experiencing oneself vs another person as being the cause of an action: the neural correlates of the experience of agency, Neuroimage 15 (2002) 596–603. [11] D.P. Friedman, E.A. Murray, Thalamic connectivity of the second somatosensory area and neighboring somatosensory fields of the lateral sulcus of the macaque, Journal of Comparative Neurology 252 (1986) 348–373. [12] D.P. Friedman, E.A. Murray, J.B. O’Neill, M. Mishkin, Cortical connections of the somatosensory fields of the lateral sulcus of macaques: evidence for a corticolimbic pathway for touch, Journal of Comparative Neurology 252 (1986) 323–347. [13] M.J. Gameroff, M. Olfson, Major depressive disorder, somatic pain, and health care costs in an urban primary care practice, Journal of Clinical Psychiatry 67 (2006) 1232–1239. [14] T. Giesecke, R.H. Gracely, D.A. Williams, M.E. Geisser, F.W. Petzke, D.J. Clauw, The relationship between depression, clinical pain, and experimental pain in a chronic pain cohort, Arthritis and Rheumatism 52 (2005) 1577–1584. [15] A. Graff-Guerrero, F. Pellicer, Y. Mendoza-Espinosa, P. Martinez-Medina, J. Romero-Romo, C. de la Fuente-Sandoval, Cerebral blood flow changes associated with experimental pain stimulation in patients with major depression, Journal of Affective Disorders 107 (2008) 161–168. [16] R.C. Kessler, W.T. Chiu, O. Demler, K.R. Merikangas, E.E. Walters, Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication, Archives of General Psychiatry 62 (2005) 617–627. [17] K. Kroenke, J. Shen, T.E. Oxman, J.W. Williams Jr., A.J. Dietrich, Impact of pain on the outcomes of depression treatment: results from the RESPECT trial, Pain 134 (2008) 209–215. [18] A.R. Laird, P.M. Fox, C.J. Price, D.C. Glahn, A.M. Uecker, J.L. Lancaster, P.E. Turkeltaub, P. Kochunov, P.T. Fox, ALE meta-analysis: controlling the false discovery rate and performing statistical contrasts, Human Brain Mapping 25 (2005) 155–164. [19] J.P. Lepine, M. Briley, The epidemiology of pain in depression, Human Psychopharmacology 19 (Suppl 1) (2004) S3–S7. [20] C.H. Lin, H.Y. Lane, C.C. Chen, S.H. Juo, C.F. Yen, Predictors of fluoxetine remission for hospitalized patients with major depressive disorder, Psychiatry and Clinical Neurosciences 65 (2011) 510–517. [21] M. Lopez-Sola, J. Pujol, R. Hernandez-Ribas, B.J. Harrison, O. ContrerasRodriguez, C. Soriano-Mas, J. Deus, H. Ortiz, J.M. Menchon, J. Vallejo, N. Cardoner, Effects of duloxetine treatment on brain response to painful stimulation in major depressive disorder, Neuropsychopharmacology 35 (2010) 2305–2317. [22] N. Makris, J.M. Goldstein, D. Kennedy, S.M. Hodge, V.S. Caviness, S.V. Faraone, M.T. Tsuang, L.J. Seidman, Decreased volume of left and total anterior insular lobule in schizophrenia, Schizophrenia Research 83 (2006) 155–171. [23] M.M. Mesulam, E.J. Mufson, Insula of the old world monkey. III: Efferent cortical output and comments on function, Journal of Comparative Neurology 212 (1982) 38–52. [24] E.J. Mufson, M.M. Mesulam, Insula of the old world monkey. II: Afferent cortical input and comments on the claustrum, Journal of Comparative Neurology 212 (1982) 23–37. [25] I. Mutschler, A. Schulze-Bonhage, V. Glauche, E. Demandt, O. Speck, T. Ball, A rapid sound-action association effect in human insular cortex, PLoS One 2 (2007) e259.
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209 [26] I. Mutschler, J. Wankerl, C. Reinbold, E. Seifritz, T. Ball, The structural basis of female empathy in the anterior insular cortex, submitted for publication. [27] I. Mutschler, B. Wieckhorst, S. Kowalevski, J. Derix, J. Wentlandt, A. SchulzeBonhage, T. Ball, Functional organization of the human anterior insular cortex, Neuroscience Letters 457 (2009) 66–70. [28] E. Peltz, F. Seifert, R. DeCol, A. Dorfler, S. Schwab, C. Maihofner, Functional connectivity of the human insular cortex during noxious and innocuous thermal stimulation, Neuroimage 54 (2011) 1324–1335. [29] P. Schweinhardt, C. Glynn, J. Brooks, H. McQuay, T. Jack, I. Chessell, C. Bountra, I. Tracey, An fMRI study of cerebral processing of brush-evoked allodynia in neuropathic pain patients, Neuroimage 32 (2006) 256–265. [30] I.A. Strigo, S.C. Matthews, A.N. Simmons, Right anterior insula hypoactivity during anticipation of homeostatic shifts in major depressive disorder, Psychosomatic Medicine 72 (2010) 316–323. [31] I.A. Strigo, A.N. Simmons, S.C. Matthews, A.D. Craig, M.P. Paulus, Association of major depressive disorder with altered functional brain response during anticipation and processing of heat pain, Archives of General Psychiatry 65 (2008) 1275–1284.
209
[32] I.A. Strigo, A.N. Simmons, S.C. Matthews, A.D. Craig, M.P. Paulus, Increased affective bias revealed using experimental graded heat stimuli in young depressed adults: evidence of emotional allodynia, Psychosomatic Medicine 70 (2008) 338–344. [33] I. Tracey, Nociceptive processing in the human brain, Current Opinion in Neurobiology 15 (2005) 478–487. [34] P.E. Turkeltaub, G.F. Eden, K.M. Jones, T.A. Zeffiro, Meta-analysis of the functional neuroanatomy of single-word reading: method and validation, Neuroimage 16 (2002) 765–780. [35] D.J. Veltman, K.J. Friston, G. Sanders, C.J. Price, Regionally specific sensitivity differences in fMRI and PET: where do they come from? Neuroimage 11 (2000) 575–588. [36] C. Wiebking, A. Bauer, M. de Greck, N.W. Duncan, C. Tempelmann, G. Northoff, Abnormal body perception and neural activity in the insula in depression: an fMRI study of the depressed “material me”, World Journal of Biology and Psychiatry 11 (2010) 538–549. [37] A.D. Craig, How do you feel? Interoception: the sense of the physiological condition of the body, Nature Reviews Neuroscience 3 (2002) 655–666.
Contents lists available at SciVerse ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Review
Pain and emotion in the insular cortex: evidence for functional reorganization in major depression Isabella Mutschler a,b,∗,1 , Tonio Ball c,d,1 , Johanna Wankerl b , Irina A. Strigo a,e,f a
Department of Psychiatry, University of California San Diego, La Jolla, CA, USA Department of Psychology, Clinical Psychology and Epidemiology, University of Basel, Switzerland Bernstein Center Freiburg, University of Freiburg, Germany d Epilepsy-Center, University Medical Center Freiburg, Germany e Research Service, VA San Diego Healthcare System, San Diego, CA, USA f VA Center of Excellence for Stress and Mental Health (CESAMH), San Diego, CA, USA b c
a r t i c l e
i n f o
Article history: Received 27 December 2011 Received in revised form 27 March 2012 Accepted 30 March 2012 Keywords: Pain Emotion Depression Emotional allodynia Functional reorganization Nociception Psychophathology Meta-analysis Insula Interoception
a b s t r a c t Major Depressive Disorder (MDD) is among the top causes of disability worldwide and many patients with depression experience pain symptoms. Little is known regarding what makes depressed persons feel like they are in pain. An increasing number of neuroimaging studies show that both physical pain and depression involve the insular cortex. The present study aimed to investigate whether emotional processing in MDD patients is topologically shifted towards the insular area(s) involved in pain processing in healthy individuals. To achieve this aim, we investigated the functional organization of the insula by conducting meta-analyses of previously published neuroimaging studies on: (1) emotion in patients with MDD, (2) emotion in healthy subjects, and (3) physical pain in healthy subjects. Our results show that the dorsal part of the insula is reproducibly activated during experimental pain in healthy individuals, with multiple separate pain-related areas aligned along its dorsal border. Regions with maximal pain-related activation likelihood estimate (ALE) were located in the posterior (left) and dorsal mid-anterior insula (left and right). Furthermore, emotion-related peaks in healthy subjects were found both in its ventral (as shown in a previous meta-analysis) and dorsal anterior part. Importantly, emotion-related peaks in depressed patients were shifted to the dorsal anterior insula, where regions related to physical pain in healthy subjects are located. This shift was reflected in the observation that median z-coordinates of emotion-related responses in the left hemisphere were significantly larger in depressed patients than in healthy controls. This shift of emotion-related responses to the dorsal insula, i.e., where pain-processing takes place in healthy subjects, may play a role in “emotional allodynia” – a notion that individuals with MDD experience pain in response to stimuli that are normally not painful. © 2012 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3. 4. 5.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Search criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author at: Department of Psychiatry, University of California San Diego, La Jolla, CA, USA. Tel.: +1 858 230 4323; fax: +1 858 642 1601. E-mail address: [email protected] (I. Mutschler). 1 These authors contributed equally. 0304-3940/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neulet.2012.03.095
205 205 205 205 205 205 206 208 208 208
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
205
1. Introduction
2.2. Inclusion criteria
Major depressive disorder (MDD) is one of the most common psychiatric disorders – with a lifetime prevalence for MDD as high as 15% worldwide [16]. It is defined as a period of at least two weeks of sustained depressed mood and/or anhedonia [16]. MDD profoundly impacts the individual’s life and health in general and has been associated with a number of adverse health outcomes including a heightened subjective experience of pain [31]. When William Styron described his depression, he said: “Mysteriously and in ways that are totally remote from natural experience, the gray drizzle of horror induced by depression takes on the quality of physical pain.” Over 75% of patients with depression experience pain symptoms [19]. Conversely, 30–60% of chronic pain patients report significant depressive symptoms [3]. The presence of comorbid pain contributes significantly to unfavorable outcomes and increased cost of treatment in MDD [13]. Little is known, however, about the neuronal mechanisms that make depressed individuals feel like they are in pain. A large number of neuroimaging studies have investigated functional brain changes in depression and physical pain. Experience of pain has been associated with activation in the insula and the anterior cingulate cortex, as well as somatosensory areas and prefrontal regions – a network often referred to as “the pain matrix” [33]. A recent meta-analysis shows that responses to emotional stimuli in depression involve the insula, prefrontal cortices, the hippocampus, the amygdala, the anterior cingulate cortex, the cerebellum, the thalamus, the caudate nucleus, the putamen and the globus pallidus [9]. Together, these findings indicate that both physical pain and depression involve the insula. Recently, it has been shown that altered activity in the insula is associated with abnormal interoception [36] and pain processing in MDD [4,14,15,21,30,31]. As the insula is strongly involved both in pain and emotional processing, in the present study, we have investigated whether there are alterations in insular activation that may help explaining “emotional allodynia”– that is why somatosensory stimuli that are typically perceived as non-painful in healthy subjects can cause emotional pain in depressed patients [32]. As a possible neural correlate to this clinical observation, we put forward the hypothesis that emotional processing in depressed patients is topologically shifted toward the insular area (s) associated with physical pain in healthy individuals. Hence, this study had the aim to investigate the functional organization of the insula by conducting activation likelihood estimate (ALE) meta-analyses of previously published neuroimaging studies on: (1) emotion in individuals with MDD, (2) emotion, and (3) physical pain in healthy subjects. As has been recently shown, meta-analyses are useful tools for detecting brain responses that are reproducibly observed across functional imaging studies and for investigating the functional organization of the human insula [25,27]. In addition, for data interpretation, an ALE on studies investigating sensorimotor tasks was carried out (for more details see reference [27]). Furthermore, we provide a brief overview of the few neuroimaging studies that examined experimental pain processing in individuals with MDD.
We included studies that: (1) investigated emotional processing or experimentally induced physical pain in healthy individuals and emotional processing in patients with MDD using functional magnetic resonance imaging (fMRI) or positron emission tomography (PET); (2) reported insular peaks that lay within 5 mm of the insular border (for more details see below); (3) provided Talairach or Montreal Neurological Institute (MNI) coordinates; (4) included hand movement (e.g., button press) only if two experimental conditions assumed to have equal movement-related activity were contrasted; this was selected to rule out movement-related effects, which are known to activate a region in the anterior insular cortex (see Fig. 1 and references [25,27]); (5) examined healthy individuals that did not suffer from any neurological or psychiatric disorder; (6) examined depressed patients meeting the criteria for MDD based on the Diagnostic and Statistical Manual, fourth edition (DSM-IV [1]) or the International Classification of Diseases, tenth revision (ICD-10 [2]).
2. Methods 2.1. Search criteria A systematic literature search in large databases (Medline, PsychInfo) for neuroimaging studies published in English prior to December 2011 was conducted. The search keywords were “depressive”; “depression”; “MDD”; “pain”; “noxious stimuli”; “emotion”; “brain imaging”; “fMRI”; “PET”. In addition; we examined the references of the selected papers.
3. Data analysis Descriptive information was extracted from each article. Talairach coordinates were translated to match the MNI space using the tal2mni Matlab script from http://eeg.sourceforge.net/ doc m2html/bioelectromagnetism/tal2mni.html. An ‘activation likelihood estimate’ (ALE), given by the union of the probabilities associated with the different foci, was calculated for an area comprising the whole Y and Z extent of the insular cortex. ALE was calculated across studies such that the summed ALE of all peaks of each study was normalized to unity to ensure that studies reporting large numbers of peaks cannot disproportionally dominate the resulting ALE map. The average extent of the insular cortex was determined from the T1 multi-subject template provided with SPM5. The anatomical boundaries of the insula are described in the study by Makris et al. [22]. The spatial union of all ALE values could be portrayed as an ALE-value map describing the reproducibility of an effect within different insular subregions. Statistical significance was assessed using a (non-parametric) single-threshold permutation test as previously described [18,34]. ALE scores inferred from the distribution of activation foci from the included studies of one functional category were compared against a single critical ALE score derived from an empirical null distribution. This null distribution resulted from 10,000 permutations of an equal number of foci that were distributed randomly throughout the insular cortex. The ALE maps that resulted from the separate analyses of different functional categories are presented. The following ALE results are reported at p < 0.05 (FDR corrected): fMRI studies on pain in healthy individuals, and studies on sensorimotor tasks (Fig. 1, red and turquoise). The remaining ALE results are reported at p < 0.001 (uncorrected), because there were no significant ALE values at the FDR-corrected threshold. Data were processed using SPM5/8 alongside with in-house developed MATLAB (Version 7.0.4, MathWorks) scripts. To evaluate whether response peaks in depressed patients were shifted dorsally (i.e., where reproducible pain-related responses were found, see below), we applied a Wilcoxon rank sum test to the z-coordinates of the reported peaks, both for those in the right and left hemisphere. This test was also applied to the y-coordinates. 4. Results Eleven fMRI studies on emotion and MDD (comprising 209 patients and 20 peaks), 44 studies on emotion in healthy subjects (comprising 756 subjects and 89 peaks), and 57 studies
206
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
Fig. 1. Shows activation likelihood estimate (ALE) findings for the left (a) and (b) for the right insula and their MNI z coordinates (on the y-axis) and y coordinates (on the x-axis) related to physical pain in healthy individuals (using fMRI = red, or PET = magenta), emotional processing in healthy subjects (yellow), emotional processing in depressed individuals (green), and sensorimotor tasks (turquoise). Studies on emotion in healthy individuals entering this ALE analysis did not include insula-coordinates which were related to the measurement of peripheral physiological changes. In a recent meta-analysis responses related to peripheral physiological changes resulting from emotional experiences were located in the ventral anterior insular cortex [27]. The solid gray line indicates the mean outline of the left sagittal insula and the dashed gray line the central sulcus of the insula dividing the insula in an anterior and posterior part (see methods for more details). In the left insular cortex, pain-related maximal ALE were found in the posterior and dorsal mid-anterior insula, whereas in the right insula the maximal ALE was located in the dorsal mid-anterior insula. Insula effects related to emotional processing in depressed individuals overlapped with the pain (left) and also sensorimotor areas (right) found in healthy subjects. Hand-movement task had their ALE maxima in the posterior part of the anterior insular cortex. This region has also been implicated in sense of agency of hand movement [10]. The reported ‘sense of agency’ related response peaks overlapped with the ALE maximum related to sensorimotor tasks (for more details see reference [27]).
on physical pain in healthy individuals (comprising 690 subjects and 175 peaks) fulfilled the inclusion criteria. Studies on pain predominantly used thermal heat stimuli (n = 29). Thirteen studies used positron emission tomography (PET) and 44 studies used fMRI. Studies on emotion used PET (N = 13) or fMRI (n = 31). Seven out of 11 studies on emotion in MDD investigated unmedicated patients. Eight studies on emotion in MDD specified that the patients and the healthy controls where matched. See Supplement for detailed description of the articles. The results of the ALE meta-analyses are shown in Fig. 1. We found that pain-related maximal ALE in healthy subjects was located in the posterior and dorsal mid-anterior insula on the left and in the dorsal mid-anterior insula on the right (Fig. 1). We further found that emotion-related maximal ALE in healthy individuals – excluding activation peaks in the insula related to the measurement of peripheral physiological changes – was located in the dorsal anterior insula. Conversely, emotion-related maximal ALE in individuals with MDD was located more dorsal and overlapped with the insula region related more to pain and sensorimotor processing in healthy subjects (Fig. 1). In other words, there was a shift of emotion-related responses in MDD towards the dorsal insula, a region where pain-processing takes place in healthy individuals. Furthermore, median z-coordinates of emotion-related responses were significantly larger in the depressed patients than in the healthy controls, in the left (p = 0.009), but not significantly in the right (p = 0.505, Fig. 2) hemisphere. The observed difference in the left insula prevailed independently of the coordinate system used in the studies (and therefore, independent of possible errors in reporting the coordinate system). Median y-coordinates of emotion-related responses were not significantly different between depressed patients and healthy controls: p = 0.547 (left) and p = 0.401 (right). We found that pain-related coordinates in healthy individuals from PET studies were located more posterior within the insular region than those from the fMRI studies. An ALE analysis on pain processing in MDD patients was not possible due to limited studies available to date.
Fig. 2. Boxplots showing the distribution of z-coordinates of emotion-related responses in depressed patients and in healthy subjects. Emotion-related peaks in depressed patients were shifted to the dorsal insula: all peaks in depressed patients were found in the dorsal part of the insula (z > 0) while, in healthy subjects, peaks were found both in the dorsal and in the ventral parts of the insula. Median zcoordinates of emotion-related responses were significantly larger in the depressed patients than in the healthy controls in the left hemisphere (p = 0.009), but not in the right (p = 0.505). This shift of emotion-related responses to the dorsal insula in MDD, i.e., where pain-processing takes place, may play a role in “emotional allodynia” – a notion that individuals with MDD hurt in response to stimuli that are not painful.
5. Discussion The main goal of this study was to investigate the hypothesis that emotional processing within the insula in depressed patients is topologically shifted toward the insular area(s) associated with physical pain in healthy individuals. Our results show that, compared to healthy subjects, emotional processing within the insula in MDD is shifted to the dorsal anterior insula, to regions including those with the maximal ALE related to pain processing in
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
healthy subjects. In support of this we found that the median zcoordinates of emotion-related responses in depressed patients compared to healthy controls were significantly larger on the left and non-significantly on the right. Taken together, the observed shift of emotion-related responses in MDD to the dorsal insula, i.e., where pain processing takes place in healthy subjects, may play a role in “emotional allodynia” – a notion that individuals with MDD hurt in response to stimuli that are not painful [32]. Our results show that many insular response peaks related to emotional processing in MDD overlapped with the maximal ALE related to sensorimotor tasks. Sensorimotor tasks reproducibly activate the posterior part of the anterior insular cortex, just anterior to the central sulcus of the insula. We previously related the reproducible movement-related activity in the anterior insula to the existence of the functional insular motor area, i.e., an area with neural processing directly involved in efferent motor control [27]. This interpretation is supported by a recent study that investigated connectivity patterns of the insula and found that the insular region involved in sensorimotor tasks showed functional connections with the premotor cortex (BA 6) that is involved in motor control [5]. Alternatively, it has been suggested that the insula contains a somatotopic representation of the subjective feelings of one’s current movements as part of a representation of all feelings from the body [7]. The anterior insula has been reported to be involved in the ‘sense of agency’ of hand movement, that is, the experience of oneself being the cause of an action [10]. The reported response peaks related to the ‘sense of agency’ overlapped with the ALE maximum associated with sensorimotor tasks [27]. We found that emotion-related peaks in healthy subjects were located in the dorsal anterior insula. In a recent meta-analysis we examined insula response specifically related to peripheral physiological changes resulting from emotional experiences and found that this activity was represented in the ventral anterior insular cortex [27]. This region was also the most likely site of insular co-activation with the amygdala. In the current study, insula-coordinates associated with emotion-related peripheral physiological changes were excluded. Therefore, our findings suggest that distinct functional insula regions may be involved in different aspects of emotional processing, whereby peripheral physiological aspects of emotional processing are mapped to the ventral anterior regions, while more cognitive–evaluative aspects are mapped to the dorsal-anterior region. This interpretation is supported by the fact that in our study-sample subjects evaluated the emotional content of the presented stimuli. This conclusion is also in line with the observed functional connectivity between the dorsal anterior insula and the dorsal anterior cingulate cortex, a network that plays a crucial role in cognitive decision-making [8]. This is also consistent with earlier findings in monkeys show that the insula is interconnected with the cingulate cortex [23]. We found that pain-related peaks in healthy individuals were located in the left posterior and dorsal mid-anterior insula, whereas in the right insula pain-related representation was located in the dorsal mid-anterior insula. Similarly, a study by Schweinhardt et al. found that experimental pain in healthy volunteers was predominantly represented in the dorsal anterior and in the posterior insula [29]. The authors suggested, that posterior insula activation in experimental pain reflect basic sensory aspects of nociceptive input, which is consistent with the animal work and the proposed role of this region as “interoceptive sensory cortex” [37]. Furthermore, we found that within the right dorsal mid-anterior insula pain, emotion, and sensorimotor related responses overlap. Our findings are thus in direct agreement with the notion that this insula region plays an integrative role by
207
bringing together information for emotional–cognitive evaluation of a noxious stimuli and the associated motor reactions. Anatomical studies in monkeys show that the anterior insula – with its reciprocal connections to the cingulate cortex, the primary somatosensory cortex, and the prefrontal cortex – is in an ideal position to associate sensory events with emotional–cognitive and motor responses during the experience of pain [11,12,23,24]. Recently, Peltz et al. investigated the functional connectivity of insular cortex subregions during noxious and innocuous thermal stimulation [28]. The authors found that the anterior part of the insula was more functionally connected to areas known for affective and cognitive processing (including the anterior cingulate cortex and the prefrontal cortex), whereas the posterior part was more connected with areas known for sensory-discriminative processing of noxious and somatosensory stimuli (including the primary somatosensory cortex). Together, the findings of the present study indicate multiple representations of pain in the dorsal insula, arranged along the anterior–posterior axis along the dorsal border of the insula. Furthermore, our ALE findings are also in agreement with the idea of a posterior to anterior functional gradient, assuming that sensory information about the body’s physiological state is mapped in the posterior insula and re-represented in the anterior insula where it becomes consciously accessible, enabling a general subjective affective experience [7]. Few brain imaging studies to date have examined pain processing in individuals with MDD [4,14,15,21,30,31], therefore ALE analysis could not be performed. However, despite differences in experimental design and the imaging technique used, all six studies found increased pain-related processing within the insula in MDD further conforming insula dysregulation in this disorder. We briefly describe them: Graff-Guerrero et al. [15] conducted a Single Photon Emission Computed Tomography (SPECT) study in 24 unmedicated individuals with MDD who were all free of pain symptoms. They compared brain activation during application of painful pressure stimuli to the right distal phalanx of the fifth finger versus sham stimulation and found increased activation within left anterior insula and within more posterior aspects of the bilateral insula. We have shown that in healthy individuals pain-related peaks from PET studies are located more posterior within the insular cortex than those from the fMRI studies, which is consistent with the reported regionally specific sensitivity differences between fMRI and PET described previously [35]. Therefore, it is difficult to infer how insula activation from the Graff-Guerrero et al. [15] study maps to emotional processing insular region in MDD found here. In an fMRI study by Bar et al. [4] brain activation during experimental heat pain processing (right volar wrist) was compared between 13 women with MDD who were all unmedicated and free of pain symptoms and 13 healthy control women. The authors found increased brain response in MDD within dorsal anterior insula on the right and more medial ventral insula on the left. Strigo et al. [31] conducted an fMRI study comparing heat pain anticipation and processing (left volar forearm) in 15 subjects with MDD who were all unmedicated and free of pain symptoms to 15 matched healthy comparison subjects who never had MDD. The authors found increased activation during pain anticipation within bilateral dorsal anterior insula in MDD. Similarly, Strigo et al. [30] found right anterior insula dysfunction in MDD. Lopez-Sola et al. [21] conducted an fMRI study comparing heat pain processing (right volar forearm) in 13 unmedicated subjects with MDD who experienced low levels of baseline pain to 20 matched healthy subjects. The authors found significantly increased activation to pain in the MDD group within more posterior aspects of the bilateral insula. Finally, Giesecke et al. [14] examined pressure pain processing (left thumb) between 53 patients with fibromyalgia with significant depressive symptoms and 42 healthy control subjects. They found
208
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209
that increased activation within dorsal anterior and more posterior insula on the right was uniquely associated with depression in their sample. Together, these studies clearly show that insula plays an important role during pain processing in individuals with MDD. It would be important to examine how pain-related activation in MDD maps to functional organization within insula cortex reported here when more studies become available. Several important limitations should be noted. Four studies on emotion and MDD investigated patients taking antidepressants, which are known to influence insula function [9]. Thus, our findings should be replicated when more studies on emotional processing in unmedicated MDD subjects become available. Furthermore, future studies should directly compare different ALE maps and examine whether there are significant laterality effects; we are beginning to address these questions [26]. Specifically, the dorsal shift in the emotion-related activation peak in MDD was significant on the left but not on the right side in our study. Interestingly, although non-significant on both sides, a visually apparent trend in the antero-posterior direction was noted in MDD on the right. It will be important to examine whether functional reorganization within the insula in MDD is lateralized, consistent with emotional asymmetry theories proposed by others [6]. Another relevant topic for further investigations would be the examination of the relationship between functionally defined areas – as investigated in the present study – and macro-anatomical landmarks of the human insula. To our knowledge no study has been published to date on the cytoarchitectonic arrangement of the human anterior insula. Finally, an ALE analysis on experimentally induced pain in MDD was not possible because to our knowledge, only six neuroimaging studies have been published to date (see above). To summarize, although pain is not a diagnostic symptom of MDD according to DSM-IV [1], it should be taken into account because over 75% of patients with depression experience pain symptoms [19]. Pain in depression has a strong negative impact on the clinical course, treatment response [17], and remission [20]. By applying a coordinate-based meta-analytical approach we found that emotion-related peaks in depressed individuals were shifted to the dorsal insula, towards the activation peaks associated with pain processing in healthy subjects. To our knowledge this is the first study which finds evidence for functional reorganization of the insula in depressed individuals, which may contribute to the neurobiological understanding of “emotional allodynia” – a notion when typically non-painful stimuli can cause emotional pain in depressed patients [32]. Does emotional pain physically hurt if one is depressed? The overlap between emotional and pain-processing circuitry found in the current study in people with MDD supports this notion. Future studies should examine whether such functional shift within the insula in MDD can be experimentally replicated and whether a similar effect might exist in other brain regions. It will be important to investigate whether this shift in emotional processing circuitry towards pain-processing circuitry within the insula is a vulnerability factor for the development of psychiatric and/or psychosomatic disorders and whether it remains after successful treatment. Conversely, it would be interesting to see whether segregation of pain-related and emotion-related activation peaks within the insula may indicate resilience for the development of psychopathology.
Acknowledgements We would like to thank Dr. Bud Craig for his advice on insula anatomy. We gratefully acknowledge the financial support by the Swiss National Science Foundation (SNF fellowship PA00P1 136408 to IM) and the National Institute of Mental Health (MH80003 to IAS).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.neulet.2012.03.095.
References [1] Diagnostic and statistical manual of mental disorders, 4th ed., American Psychiatric Association, Washington, 1994. [2] International Classification of Diseases (ICD), 10th revision, World Health Organization, 1996. [3] M.J. Bair, R.L. Robinson, W. Katon, K. Kroenke, Depression and pain comorbidity: a literature review, Archives of Internal Medicine 163 (2003) 2433–2445. [4] K.J. Bar, G. Wagner, M. Koschke, S. Boettger, M.K. Boettger, R. Schlosser, H. Sauer, Increased prefrontal activation during pain perception in major depression, Biological Psychiatry 62 (2007) 1281–1287. [5] L. Cerliani, R.M. Thomas, S. Jbabdi, J.C. Siero, L. Nanetti, A. Crippa, V. Gazzola, H. D’Arceuil, C. Keysers, Probabilistic tractography recovers a rostrocaudal trajectory of connectivity variability in the human insular cortex, Human Brain Mapping (2011) [Epub ahead of print]. [6] A.D. Craig, Forebrain emotional asymmetry: a neuroanatomical basis? Trends in Cognitive Science 9 (2005) 566–571. [7] A.D. Craig, How do you feel – now? The anterior insula and human awareness, Nature Reviews. Neuroscience 10 (2009) 59–70. [8] B. Deen, N.B. Pitskel, K.A. Pelphrey, Three systems of insular functional connectivity identified with cluster analysis, Cerebral Cortex 21 (2011) 1498–1506. [9] P. Delaveau, M. Jabourian, C. Lemogne, S. Guionnet, L. Bergouignan, P. Fossati, Brain effects of antidepressants in major depression: a meta-analysis of emotional processing studies, Journal of Affective Disorders 130 (2011) 66–74. [10] C. Farrer, C.D. Frith, Experiencing oneself vs another person as being the cause of an action: the neural correlates of the experience of agency, Neuroimage 15 (2002) 596–603. [11] D.P. Friedman, E.A. Murray, Thalamic connectivity of the second somatosensory area and neighboring somatosensory fields of the lateral sulcus of the macaque, Journal of Comparative Neurology 252 (1986) 348–373. [12] D.P. Friedman, E.A. Murray, J.B. O’Neill, M. Mishkin, Cortical connections of the somatosensory fields of the lateral sulcus of macaques: evidence for a corticolimbic pathway for touch, Journal of Comparative Neurology 252 (1986) 323–347. [13] M.J. Gameroff, M. Olfson, Major depressive disorder, somatic pain, and health care costs in an urban primary care practice, Journal of Clinical Psychiatry 67 (2006) 1232–1239. [14] T. Giesecke, R.H. Gracely, D.A. Williams, M.E. Geisser, F.W. Petzke, D.J. Clauw, The relationship between depression, clinical pain, and experimental pain in a chronic pain cohort, Arthritis and Rheumatism 52 (2005) 1577–1584. [15] A. Graff-Guerrero, F. Pellicer, Y. Mendoza-Espinosa, P. Martinez-Medina, J. Romero-Romo, C. de la Fuente-Sandoval, Cerebral blood flow changes associated with experimental pain stimulation in patients with major depression, Journal of Affective Disorders 107 (2008) 161–168. [16] R.C. Kessler, W.T. Chiu, O. Demler, K.R. Merikangas, E.E. Walters, Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication, Archives of General Psychiatry 62 (2005) 617–627. [17] K. Kroenke, J. Shen, T.E. Oxman, J.W. Williams Jr., A.J. Dietrich, Impact of pain on the outcomes of depression treatment: results from the RESPECT trial, Pain 134 (2008) 209–215. [18] A.R. Laird, P.M. Fox, C.J. Price, D.C. Glahn, A.M. Uecker, J.L. Lancaster, P.E. Turkeltaub, P. Kochunov, P.T. Fox, ALE meta-analysis: controlling the false discovery rate and performing statistical contrasts, Human Brain Mapping 25 (2005) 155–164. [19] J.P. Lepine, M. Briley, The epidemiology of pain in depression, Human Psychopharmacology 19 (Suppl 1) (2004) S3–S7. [20] C.H. Lin, H.Y. Lane, C.C. Chen, S.H. Juo, C.F. Yen, Predictors of fluoxetine remission for hospitalized patients with major depressive disorder, Psychiatry and Clinical Neurosciences 65 (2011) 510–517. [21] M. Lopez-Sola, J. Pujol, R. Hernandez-Ribas, B.J. Harrison, O. ContrerasRodriguez, C. Soriano-Mas, J. Deus, H. Ortiz, J.M. Menchon, J. Vallejo, N. Cardoner, Effects of duloxetine treatment on brain response to painful stimulation in major depressive disorder, Neuropsychopharmacology 35 (2010) 2305–2317. [22] N. Makris, J.M. Goldstein, D. Kennedy, S.M. Hodge, V.S. Caviness, S.V. Faraone, M.T. Tsuang, L.J. Seidman, Decreased volume of left and total anterior insular lobule in schizophrenia, Schizophrenia Research 83 (2006) 155–171. [23] M.M. Mesulam, E.J. Mufson, Insula of the old world monkey. III: Efferent cortical output and comments on function, Journal of Comparative Neurology 212 (1982) 38–52. [24] E.J. Mufson, M.M. Mesulam, Insula of the old world monkey. II: Afferent cortical input and comments on the claustrum, Journal of Comparative Neurology 212 (1982) 23–37. [25] I. Mutschler, A. Schulze-Bonhage, V. Glauche, E. Demandt, O. Speck, T. Ball, A rapid sound-action association effect in human insular cortex, PLoS One 2 (2007) e259.
I. Mutschler et al. / Neuroscience Letters 520 (2012) 204–209 [26] I. Mutschler, J. Wankerl, C. Reinbold, E. Seifritz, T. Ball, The structural basis of female empathy in the anterior insular cortex, submitted for publication. [27] I. Mutschler, B. Wieckhorst, S. Kowalevski, J. Derix, J. Wentlandt, A. SchulzeBonhage, T. Ball, Functional organization of the human anterior insular cortex, Neuroscience Letters 457 (2009) 66–70. [28] E. Peltz, F. Seifert, R. DeCol, A. Dorfler, S. Schwab, C. Maihofner, Functional connectivity of the human insular cortex during noxious and innocuous thermal stimulation, Neuroimage 54 (2011) 1324–1335. [29] P. Schweinhardt, C. Glynn, J. Brooks, H. McQuay, T. Jack, I. Chessell, C. Bountra, I. Tracey, An fMRI study of cerebral processing of brush-evoked allodynia in neuropathic pain patients, Neuroimage 32 (2006) 256–265. [30] I.A. Strigo, S.C. Matthews, A.N. Simmons, Right anterior insula hypoactivity during anticipation of homeostatic shifts in major depressive disorder, Psychosomatic Medicine 72 (2010) 316–323. [31] I.A. Strigo, A.N. Simmons, S.C. Matthews, A.D. Craig, M.P. Paulus, Association of major depressive disorder with altered functional brain response during anticipation and processing of heat pain, Archives of General Psychiatry 65 (2008) 1275–1284.
209
[32] I.A. Strigo, A.N. Simmons, S.C. Matthews, A.D. Craig, M.P. Paulus, Increased affective bias revealed using experimental graded heat stimuli in young depressed adults: evidence of emotional allodynia, Psychosomatic Medicine 70 (2008) 338–344. [33] I. Tracey, Nociceptive processing in the human brain, Current Opinion in Neurobiology 15 (2005) 478–487. [34] P.E. Turkeltaub, G.F. Eden, K.M. Jones, T.A. Zeffiro, Meta-analysis of the functional neuroanatomy of single-word reading: method and validation, Neuroimage 16 (2002) 765–780. [35] D.J. Veltman, K.J. Friston, G. Sanders, C.J. Price, Regionally specific sensitivity differences in fMRI and PET: where do they come from? Neuroimage 11 (2000) 575–588. [36] C. Wiebking, A. Bauer, M. de Greck, N.W. Duncan, C. Tempelmann, G. Northoff, Abnormal body perception and neural activity in the insula in depression: an fMRI study of the depressed “material me”, World Journal of Biology and Psychiatry 11 (2010) 538–549. [37] A.D. Craig, How do you feel? Interoception: the sense of the physiological condition of the body, Nature Reviews Neuroscience 3 (2002) 655–666.