Leerssen et al. Translational Psychiatry (2020)10:425 https://doi.org/10.1038/s41398-020-01109-5 Translational Psychiatry ARTICLE Open Access Brain structural correlates of insomnia severity in 1053 individuals with major depressive disorder: results from the ENIGMA MDD Working Group Jeanne Leerssen 1,2, Tessa F. Blanken 1,2, Elena Pozzi3,4, Neda Jahanshad 5, Lyubomir Aftanas6,7, Ole A. Andreassen8,9, Bernhard T. Baune10,11, Ivan Brack 6, Angela Carballedo12, Christopher R. K. Ching5, Udo Dannlowski11, Katharina Dohm11, Verena Enneking11, Elena Filimonova6, Stella M. Fingas11, Thomas Frodl 12,13, Beata R. Godlewska14, Janik Goltermann11, Ian H. Gotlib 15, Dominik Grotegerd11, Oliver Gruber16, Mathew A. Harris 17, Sean N. Hatton18, Emma Hawkins17, Ian B. Hickie18, Natalia Jaworska 19,20, Tilo Kircher21, Axel Krug 21,22, Jim Lagopoulos23, Hannah Lemke11, Meng Li24,25, Frank P. MacMaster26,27, Andrew M. McIntosh 17,28, Quinn McLellan 26,29, Susanne Meinert11, Benson Mwangi30, Igor Nenadić21, Evgeny Osipov 6, Maria J. Portella31,32, Ronny Redlich11,33, Jonathan Repple11, Matthew D. Sacchet34, Philipp G. Sämann35, Egle Simulionyte16, Jair C. Soares36, Martin Walter24,37, Norio Watanabe38, Heather C. Whalley 17, Dilara Yüksel21,39, Dick J. Veltman40,41, Paul M. Thompson5, Lianne Schmaal 4,42 and Eus J. W. Van Someren1,2,40 Abstract 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; It has been difficult to find robust brain structural correlates of the overall severity of major depressive disorder (MDD). We hypothesized that specific symptoms may better reveal correlates and investigated this for the severity of insomnia, both a key symptom and a modifiable major risk factor of MDD. Cortical thickness, surface area and subcortical volumes were assessed from T1-weighted brain magnetic resonance imaging (MRI) scans of 1053 MDD patients (age range 13-79 years) from 15 cohorts within the ENIGMA MDD Working Group. Insomnia severity was measured by summing the insomnia items of the Hamilton Depression Rating Scale (HDRS). Symptom specificity was evaluated with correlates of overall depression severity. Disease specificity was evaluated in two independent samples comprising 2108 healthy controls, and in 260 clinical controls with bipolar disorder. Results showed that MDD patients with more severe insomnia had a smaller cortical surface area, mostly driven by the right insula, left inferior frontal gyrus pars triangularis, left frontal pole, right superior parietal cortex, right medial orbitofrontal cortex, and right supramarginal gyrus. Associations were specific for insomnia severity, and were not found for overall depression severity. Associations were also specific to MDD; healthy controls and clinical controls showed differential insomnia severity association profiles. The findings indicate that MDD patients with more severe insomnia show smaller surfaces in several frontoparietal cortical areas. While explained variance remains small, symptom-specific associations could bring us closer to clues on underlying biological phenomena of MDD. Introduction Correspondence: Jeanne Leerssen (j.leerssen@vu.nl) 1 Multiple findings highlight the importance of insomnia Department of Sleep and Cognition, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, for psychiatric disorders in general, and in particular for Amsterdam, The Netherlands 2 major depressive disorder (MDD)1. Insomnia is a primary Department of Integrative Neurophysiology, Center for Neurogenomics and risk factor for developing MDD, e.g., ref. 2, and its presence Cognitive Research (CNCR), Amsterdam Neuroscience, VU University Amsterdam, Amsterdam, The Netherlands in people suffering from MDD hampers the effectiveness of Full list of author information is available at the end of the article © The Author(s) 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Leerssen et al. Translational Psychiatry (2020)10:425 Page 2 of 11 clinical interventions, e.g., ref. 3. Treating insomnia can also NeuroImaging Genetics through Mega-Analysis) MDD improve the outcome of patients suffering from depres- working group (http://enigma.ini.usc.edu/). We included sion4,5. Moreover, recent genome-wide association studies 1053 people who met criteria for current MDD and had report a strong genetic correlation between insomnia and completed the Hamilton Depression Rating Scale depressive symptoms and MDD6,7. Given these findings, it (HDRS)24. Supplementary Table S1 lists the diagnostic seems highly relevant to identify neural correlates of instruments and the exclusion criteria applied at each of insomnia severity in people suffering from MDD. the 15 participating sites. Additional data from clinical To date, brain structural correlates of insomnia symptoms controls and healthy controls were assembled to evaluate in people with MDD are largely unexplored. Elucidating such whether insomnia associations were specific to MDD (see correlates may provide key clues to ultimately uncovering the Supplementary Methods for details). For clinical controls, neural correlates of the risk for MDD development. Several we were able to include 260 patients from 5 ENIGMA BD anatomical magnetic resonance imaging studies compared working group sites in whom the HDRS had been asses- people with insomnia disorder (ID) without MDD to those sed (see Supplementary Table S2 for demographics). Next without sleep complaints. People with ID reported smaller to a first healthy control sample of ENIGMA (n = 1277 gray matter (GM) volumes in the orbitofrontal (OFC)8–10, completed the HDRS), we evaluated associations in a parietal8 as well as middle cingulate11 cortices, the pineal second healthy control sample from the Human Con- gland12, the thalamus13, and a smaller volume and surface nectome Project (HCP)25 of which n = 831 had completed area in the inferior frontal gyrus pars triangularis14, as well as the Pittsburgh Sleep Quality Index (PSQI)26 (see Supple- a larger GM volume in the rostral anterior cingulate cortex mentary Table S3 for demographics). Exclusion criteria (rACC)9. Some studies have suggested a smaller hippo- for healthy controls were a history of MDD, a current campal volume in people with insomnia15,16, but these diagnosis of MDD, or any other psychiatric disorders. All findings could not be replicated, e.g., refs. 8–10,17. Other sites obtained approval from local institutional review studies in people with ID assessed cortical thickness and boards and ethics committees. All participants provided found a thinner cortex in the ACC, precentral and lateral informed consent. prefrontal cortex18 and a thicker cortex in several OFC regions, the rACC, middle cingulate cortex, insula, superior Severity of insomnia and overall depression severity parietal cortex, and fusiform area19. In MDD patients, brain Three HDRS items were summed to obtain a valid structural correlates of insomnia severity have hardly been insomnia severity score27 and the remaining items for an investigated. A larger amygdala and smaller medial OFC insomnia-independent depression severity score, here refer- have been reported in MDD patients with insomnia20,21 as red to as the HDRS-14 (Supplementary Methods). In the compared to MDD patients without insomnia. second healthy control sample (HCP), corresponding PSQI It is tempting to presume that brain areas involved in the items were summed to obtain an insomnia severity score and severity of insomnia in people without MDD are also an insomnia-independent depression severity score was cal- involved in the severity of insomnia in people suffering from culated by excluding the sleep item from the total depression MDD. However, the complexity of the neuronal networks score of the Achenbach Adult Self Report questionnaire28. involved in sleep regulation and MDD makes it also con- Supplementary Methods provide details and validation. ceivable that different brain mechanisms can underlie see- mingly similar sleep complaints22,23. The present study Image processing and analysis therefore applied a whole-brain analysis to uncover brain Image acquisition parameters for each site are provided structural correlates of insomnia severity in people diag- in Supplementary Table S4. Schmaal et al. and Glasser nosed with MDD. We evaluated, in a sample of 1053 MDD et al.29–31 provide details of the use of FreeSurfer32 seg- patients, whether insomnia severity was associated with mentation to obtain surface area and thickness of 68 global and regional differences in cortical thickness, cortical cortical regions33, as well as 14 subcortical volumes, lat- surface areas, and volumes of subcortical regions. Addi- eral ventricle volumes, 2 whole-hemisphere measures, and tionally, we evaluated whether the identified associations: (1) intracranial volume (ICV). were specific to insomnia or driven by overall depression severity and (2) specific to MDD or also present in healthy Statistical analyses controls (n = 2108) and clinical controls with bipolar dis- MDD patients order (BD; n = 260). Linear mixed-effects models regressed insomnia severity on surface and thickness of cortical regions and subcortical Materials and methods volume. First, we evaluated whether insomnia severity could Samples be predicted from the overall cortical surface area, its aver- Data for the main analysis were assembled from 15 age thickness, or from total subcortical volume. Separate independent samples of the ENIGMA (Enhancing models subsequently evaluated individual brain regions. Leerssen et al. Translational Psychiatry (2020)10:425 Page 3 of 11 Models were adjusted for age, sex, scanner site (random Results intercept), insomnia-independent depression severity, and, MDD patients for subcortical volumes, total ICV. False discovery rate Table 1 summarizes the characteristics of the MDD (FDR)34 correction (p < 0.05) was applied to correct the p patients included at each site separately and overall. values for multiple comparisons for cortical surface areas Linear mixed-effects regression indicated more severe and thickness, and subcortical volumes (respectively, 68, 68, insomnia in cases with a smaller total cortical surface area and 16 comparisons). (f2 = 0.01, ΔR2 = 0.9%, p = 0.044). Table 2, Fig. 1, and Specificity of detected associations for insomnia versus Supplementary Fig. S1 provide the results from sub- overall depressive symptoms severity was assessed with sequent mixed-effects regression analyses to investigate corresponding models with either overall HDRS-17 which cortical parcels contributed most to this inverse depression severity or the HDRS-14 insomnia-indepen- association. In brief, MDD patients with more severe dent depression severity as outcome. insomnia had smaller surface areas of the right insula Ancillary mixed-effects models including interaction (f2 = 0.02, ΔR2 = 1.5%, pcorrected = 0.031), left inferior terms (e.g., surface area * age, surface area * sex, surface frontal gyrus pars triangularis (f2 = 0.02, ΔR2 = 1.8%, area * antidepressant use, surface area * depression pcorrected = 0.018), the left frontal pole (f2 = 0.01, ΔR2 = recurrence, surface area * age of onset of depression) 0.6%, pcorrected = 0.031), right superior parietal cortex investigated whether the association of insomnia severity (f2 = 0.02, ΔR2 = 1.6%, pcorrected = 0.026), right medial with cortical surface area, thickness, or subcortical OFC (f2 = 0.02, ΔR2 = 1.3%, pcorrected = 0.031), and the volume was modified or confounded by age, sex, the use right supramarginal gyrus (f2 = 0.02, ΔR2 = 1.3%, of antidepressant medication, depression recurrence (first pcorrected = 0.031) (Fig. 1 and Table 2). Together, these versus recurrent episode patients), or age of onset of brain regions explained 2.7% of the variance in insomnia, depression. above and beyond the variance explained by the covari- To obtain effect size measures for single regressors ates. Models including additional covariates (anti- within multivariable mixed-effects models, we calculated depressant medication, depression recurrence, age of Cohen’s f2 statistic. Values of 0.02, 0.15, and 0.35, onset of depression) did not change the association respectively, indicate a small, medium, or large effect. between surface area and insomnia severity (see Supple- The proportion of variance in insomnia severity mentary Results). Ancillary analyses showed that the uniquely explained by the significant brain regions (ΔR2) association between surface area and insomnia severity above and beyond the covariates (age, sex, HDRS-14, site, was not modified or confounded by sex, use of anti- and ICV for subcortical areas) was computed by sub- depressant medication, depression recurrence, or age of tracting the explained variance of a model with only the onset of depression. A significant interaction was found covariates from the explained variance of the full model between total surface area and age (p = 0.046) (see Sup- (brain area, age, sex, HDRS-14, site, and ICV for sub- plementary Results). The surface area regions we found cortical areas) using the MuMln package in R (www.R- explain more variance in insomnia severity than they project.org). explain variance in overall depression severity (see Sup- plemental Results and Supplementary Table S5). Insomnia severity was not associated with average (p = Clinical controls and healthy controls 0.174) or regional cortical thickness (all pcorrected > 0.574), Within each of the control samples (BD clinical controls nor with total (p = 0.595) or local subcortical volume (all and healthy controls), mixed-effects analyses were repe- pcorrected > 0.886; see Supplementary Tables S6 and S7). ated, including the same covariates and FDR correction. Linear mixed-effects regression models with overall To formally evaluate whether the insomnia-related brain depression severity (HDRS-17) or adjusted depression associations found in MDD were similar or different severity (HDRS-14) as outcome measures revealed no sig- compared to each of the control samples, models nificant predictive value of total (all p > 0.300) or regional including an interaction term were additionally per- surface area (all pcorrected > 0.608), nor for overall average formed, e.g., surface area * disorder (MDD versus BD). (all p > 0.568) or local cortical thickness (all pcorrected > Interaction analyses may lack power and require a larger 0.810), or total (p > 0.354) or local subcortical volume (all sample size35. Therefore, it was additionally evaluated pcorrected > 0.238). whether adding controls to the ENIGMA MDD sample would alter the effect sizes we found in MDD patients. An Clinical controls and healthy controls increase in effect size would support a similar or even In BD clinical controls, insomnia severity was not sig- stronger association in controls as found in MDD. A nificantly associated with any of the six surface areas decrease in effect size on the contrary would suggest that found in MDD (all f2 < 0.01, p > 0.205), neither with any of controls only add noise or have an opposite association. the other local surface area, thickness, or subcortical Table 1 Demographics and clinical characteristics of current MDD patients. Study Sample N Age % Male % antidepressant % First episode Age of onset MDD HDRS-17 HDRS-14 HDRS Insomnia users MDD/recurrent Mean SD MDD Mean SD Mean SD Mean SD Mean SD 1 Barcelona 35 47.0 7.1 26 91 51/49 35.1 10.8 19.6 4.5 17.0 4.2 2.6 1.6 2 Calgary 29 17.7 1.8 52 69 0/100 14.1 2.0 20.6 6.8 18.1 6.1 2.6 1.8 3 CLING 27 34.6 11.8 56 100 67/33 31.4 10.6 20.0 4.3 16.7 3.7 3.3 1.4 4 Dublin 52 41.6 10.8 37 71 15/85 25.3 12.8 23.6 5.0 19.6 4.8 4.0 1.5 a a a 5 Edinburgh 18 22.9 3.1 33 22 NA 21.4 3.3 5.7 5.8 4.4 4.7 1.3 1.6 6 FOR2017—Marburg 192 37.3 13.5 41 67 17/83 26.5 13.0 9.8 6.3 8.2 5.4 1.6 1.6 7 FOR2017—Münster 29 35.1 13.9 45 83 24/76 25.6 12.8 11.9 7.4 10.0 6.4 1.9 1.6 8 Houston 56 38.6 13.1 27 2 25/71 21.4 11.2 11.7 7.9 9.6 6.5 2.1 2.0 Leerssen et al. Translational Psychiatry (2020)10:425 9 Magdeburg 14 39.5 12.0 71 100 21/79 31.6 11.9 12.5 4.0 10.1 4.0 2.4 2.0 10 MPIP 225 46.0 14.0 51 88 34/66 35.7 14.9 24.3 6.2 20.4 5.3 3.9 2.0 11 Münster 184 37.3 12.0 45 90 23/77 29.1 12.1 19.0 4.2 15.8 3.6 3.2 1.8 Neuroimaging Cohort 12 Novosibirsk 67 44.7 12.8 27 34 52/48 37.2 14.0 19.0 5.6 16.1 5.1 2.9 1.7 13 Oxford 38 30.1 10.6 37 0 50/50 25.6 9.1 23.0 4.1 18.7 3.6 4.3 1.4 14 Stanford 50 37.2 10.2 42 48 8/88 19.2 8.7 14.6 5.9 12.6 5.2 1.9 1.4 15 Sydney 37 23.2 12.7 19 76 27/73 15.3 6.4 21.6 5.6 18.4 4.5 3.2 1.8 Total 1053 38.6 14.0 41 69 27/71 28.6 13.8 17.9 7.9 15.0 6.8 2.9 2.0 MDD major depressive disorder, HDRS Hamilton Depression Rating Scale, NA not measured. a HDRS scores may be low because they have not been assessed simultaneously with the diagnosis. Page 4 of 11 Leerssen et al. Translational Psychiatry (2020)10:425 Page 5 of 11 Table 2 Mixed effect regression analyses estimates of the association of insomnia severity with cortical surface areas (HDRS points/cm2) in MDD patients, adjusted for age, sex, insomnia-independent depression severitya, and scanning site. B s.e. 95% CI t-value p value FDR p value N Left inferior frontal gyrus pars triangularis −0.10 0.03 −0.16 to −0.05 −3.66 0.000 0.018 1032 Right superior parietal cortex −0.03 0.01 −0.05 to −0.01 −3.37 0.001 0.026 1032 Left frontal pole −0.49 0.15 −0.79 to −0.19 −3.17 0.002 0.031 1051 Right medial orbitofrontal cortex −0.08 0.03 −0.13 to −0.03 −3.09 0.002 0.031 1032 Right supramarginal gyrus −0.03 0.01 −0.06 to −0.01 −3.03 0.003 0.031 969 Right insula −0.06 0.02 −0.10 to −0.02 −3.00 0.003 0.031 1019 Right inferior frontal gyrus pars triangularis −0.06 0.02 −0.11 to −0.02 −2.73 0.006 0.062 1020 Left insula −0.06 0.02 −0.10 to −0.01 −2.63 0.009 0.073 1029 Left superior parietal cortex −0.02 0.01 −0.04 to 0.00 −2.42 0.016 0.118 1027 Right frontal pole −0.29 0.12 −0.53 to −0.05 −2.36 0.018 0.124 1050 Right paracentral lobule −0.06 0.03 −0.11 to −0.01 −2.22 0.027 0.167 1046 Left entorhinal cortex −0.16 0.07 −0.30 to −0.01 −2.14 0.033 0.184 848 Right parahippocampal gyrus −0.12 0.06 −0.23 to 0.00 −2.05 0.041 0.214 1038 Left parahippocampal gyrus −0.11 0.06 −0.22 to 0.00 −1.99 0.047 0.227 1036 Right postcentral gyrus −0.02 0.01 −0.04 to 0.00 −1.95 0.052 0.234 1033 Left posterior cingulate cortex −0.06 0.03 −0.12 to 0.00 −1.92 0.055 0.234 1046 Right precentral gyrus −0.02 0.01 −0.04 to 0.00 −1.89 0.059 0.237 1043 Right inferior frontal gyrus pars opercularis −0.04 0.02 −0.09 to 0.00 −1.84 0.065 0.247 1020 Right superior frontal gyrus −0.01 0.01 −0.03 to 0.00 −1.75 0.080 0.275 1044 Left precentral gyrus −0.02 0.01 −0.04 to 0.00 −1.72 0.085 0.275 1033 Right inferior temporal gyrus −0.02 0.01 −0.04 to 0.00 −1.71 0.087 0.275 1026 Right entorhinal cortex −0.13 0.08 −0.28 to 0.02 −1.70 0.089 0.275 822 Left inferior frontal gyrus pars opercularis −0.03 0.02 −0.08 to 0.01 −1.60 0.111 0.328 1034 Right transverse temporal gyrus −0.14 0.09 −0.32 to 0.04 −1.53 0.127 0.360 1050 Right middle temporal gyrus −0.02 0.01 −0.04 to 0.01 −1.48 0.139 0.377 1007 Right fusiform gyrus −0.02 0.01 −0.05 to 0.01 −1.45 0.148 0.384 1030 Left superior frontal gyrus −0.01 0.01 −0.02 to 0.00 −1.43 0.153 0.384 1034 Left fusiform gyrus −0.02 0.01 −0.05 to 0.01 −1.41 0.158 0.385 1043 Right precuneus −0.01 0.01 −0.04 to 0.01 −1.29 0.199 0.440 1046 Left postcentral gyrus −0.01 0.01 −0.04 to 0.01 −1.25 0.213 0.440 1028 Left supramarginal gyrus −0.01 0.01 −0.04 to 0.01 −1.23 0.218 0.440 959 Left rostral anterior cingulate cortex −0.04 0.04 −0.12 to 0.03 −1.23 0.219 0.440 1032 Left inferior frontal gyrus pars orbitalis −0.08 0.07 −0.22 to 0.05 −1.22 0.224 0.440 1041 Left rostral middle frontal gyrus −0.01 0.01 −0.02 to 0.01 −1.21 0.225 0.440 1026 Left middle temporal gyrus −0.02 0.01 −0.05 to 0.01 −1.21 0.228 0.440 978 Right isthmus cingulate cortex 0.04 0.03 −0.03 to 0.11 1.19 0.233 0.440 1049 Left lingual gyrus −0.02 0.01 −0.04 to 0.01 −1.17 0.241 0.442 1045 Right temporal pole −0.10 0.09 −0.27 to 0.07 −1.15 0.252 0.443 1029 Left paracentral lobule −0.03 0.03 −0.09 to 0.02 −1.13 0.257 0.443 1047 Leerssen et al. Translational Psychiatry (2020)10:425 Page 6 of 11 Table 2 continued B s.e. 95% CI t-value p value FDR p value N Right banks superior temporal sulcus 0.05 0.04 −0.03 to 0.13 1.12 0.263 0.443 990 Right rostral middle frontal gyrus −0.01 0.01 −0.02 to 0.01 −1.11 0.267 0.443 1030 Right lateral occipital cortex −0.01 0.01 −0.03 to 0.01 −1.05 0.295 0.478 1037 Right superior temporal gyrus −0.02 0.01 −0.04 to 0.01 −1.03 0.304 0.480 915 Left medial orbitofrontal cortex −0.02 0.02 −0.06 to 0.02 −0.97 0.332 0.513 1023 Left isthmus cingulate cortex −0.03 0.03 −0.09 to 0.03 −0.95 0.342 0.516 1043 Left lateral occipital cortex −0.01 0.01 −0.03 to 0.01 −0.94 0.350 0.517 1037 Right rostral anterior cingulate cortex −0.03 0.04 −0.11 to 0.04 −0.91 0.362 0.524 1039 Left inferior temporal gyrus −0.01 0.01 −0.03 to 0.01 −0.86 0.388 0.549 1009 Right lingual gyrus −0.01 0.01 −0.04 to 0.02 −0.81 0.420 0.580 1037 Left transverse temporal gyrus −0.06 0.07 −0.20 to 0.08 −0.80 0.427 0.580 1050 Left inferior parietal cortex −0.01 0.01 −0.02 to 0.01 −0.74 0.460 0.614 1011 Left pericalcarine cortex −0.02 0.02 −0.06 to 0.03 −0.71 0.475 0.620 1013 Left precuneus −0.01 0.01 −0.03 to 0.02 −0.69 0.491 0.620 1042 Left caudal anterior cingulate cortex −0.03 0.04 −0.10 to 0.05 −0.69 0.492 0.620 1045 Right inferior parietal cortex 0.01 0.01 −0.01 to 0.02 0.64 0.520 0.642 1019 Right caudal anterior cingulate cortex −0.02 0.03 −0.09 to 0.05 −0.61 0.542 0.650 1042 Left superior temporal gyrus −0.01 0.01 −0.04 to 0.02 −0.61 0.545 0.650 915 Right caudal middle frontal gyrus −0.01 0.01 −0.04 to 0.02 −0.59 0.555 0.651 1039 Left cuneus 0.01 0.03 −0.04 to 0.06 0.51 0.607 0.700 1010 Right inferior frontal gyrus pars orbitalis −0.02 0.06 −0.13 to 0.09 −0.34 0.736 0.834 1038 Left banks superior temporal sulcus 0.01 0.04 −0.06 to 0.08 0.31 0.755 0.842 961 Right pericalcarine cortex 0.01 0.02 −0.04 to 0.05 0.28 0.780 0.856 1010 Left caudal middle frontal gyrus 0.00 0.01 −0.03 to 0.03 −0.23 0.820 0.875 1036 Left lateral orbitofrontal cortex 0.00 0.02 −0.04 to 0.03 −0.22 0.826 0.875 1045 Right lateral orbitofrontal cortex 0.00 0.02 −0.03 to 0.04 0.21 0.836 0.875 1047 Right cuneus 0.00 0.03 −0.06 to 0.05 −0.12 0.907 0.934 1017 Right posterior cingulate cortex 0.00 0.03 −0.06 to 0.06 0.06 0.953 0.967 1048 Left temporal pole 0.00 0.09 −0.17 to 0.17 0.04 0.967 0.967 1025 MDD major depressive disorders, HDRS Hamilton Depression Rating Scale, CI confidence interval, FDR false discovery rate. a Insomnia-independent depression severity is calculated by subtracting the insomnia scores from the total HDRS score. volume measures (all pcorrected > 0.984). To formally BD patients. When combining the BD and MDD sample evaluate whether the association between insomnia (n = 1313), the effect sizes decreased by 29–71% as severity and surface areas differed between MDD and BD compared to the effects found for cortical surface area in patients, interaction analyses were performed for each of MDD only. the six surface areas found in MDD and type of disorder In the ENIGMA-MDD healthy controls, insomnia (MDD versus BD). A significant interaction effect was severity was not significantly associated with any of the six found in only 2 out of the 6 surface areas (left inferior surface areas found in MDD (all f2 < 0.01, p > 0.193), frontal gyrus pars triangularis, p = 0.022; right supra- neither with any of the other global or local surface area, marginal gyrus, p = 0.045), indicating that a smaller sur- thickness, or subcortical volume measures (all pcorrected > face in these two areas was associated with higher 0.441). When adding the ENIGMA healthy controls to the insomnia severity specifically in MDD patients but not in MDD sample (n = 2330), the effect sizes decreased by Leerssen et al. Translational Psychiatry (2020)10:425 Page 7 of 11 by the right insula, left inferior frontal gyrus pars triangu- RH laris, left frontal pole, right superior parietal cortex, right medial OFC, and right supramarginal gyrus, that all showed significant regional effects. The association was independent of depression severity adjusted for the three insomnia items, and was specific for surface area: no associations were found for cortical thickness or subcortical volumes. The associa- tion between surface area and insomnia severity seems specific to MDD patients, since no associations were found LH in healthy or clinical controls. Cortical surface area only explained a small proportion of the variance in insomnia severity, which may not be surprising, because it is con- ceivable that a variety of other factors influence the complex trait of insomnia. On the other hand, small effects in large samples are more likely to be reliable and reproducible than large effect in small samples36. We found that surface area was specifically associated Lateral View Medial View with insomnia severity, not with overall depression severity. Our meta-analysis29 in a large overlapping sam- ple of adult MDD patients and controls from ENIGMA -5 -4 -3 -2 -1 0 1 2 3 4 5 MDD reported no significant association between cortical T-score surface area and depression severity measured using the Fig. 1 T-scores for brain regions that show a significant (pcorrected total score of the HDRS. A weak negative association was < 0.05) decreased surface area associated with higher insomnia only found between self-reported depression severity severity scores in major depressive disorder patients. Models are (Beck’s Depression Inventory, BDI-II score) and surface adjusted for age, sex, insomnia-independent depression severity (HDRS-14) and site. RH right hemisphere, LF left hemisphere. areas of the bilateral precuneus, left frontal pole, and left postcentral gyrus. Our current findings indicate a better association of total and regional surface areas for the 34–64% with respect to the significant effects found for severity of a single-domain phenotype (insomnia symp- cortical surface area in MDD only. In the HCP healthy toms) than for the severity of a multi-domain phenotype controls, insomnia severity was only significantly asso- (all/other mixed symptoms of depression). It should be ciated with 1 out of the 6 surface areas found in MDD noted that the explained variance is still small, as is (right medial OFC, f2 = 0.008, p = 0.009; other regions commonly found across genetic and neuroimaging f2 < 0.002, p > 0.188). No significant association was found regressors for complex traits like insomnia and depres- for any of the other local surface areas or for subcortical sion. While the findings thus do not explain much of volumes (all pcorrected > 0.089), whereas a significant individual differences, they may bring us a bit closer to association was found for 2 out of 68 cortical thickness clues on underlying biological phenomena involved. regions. Healthy controls with more severe insomnia The Research Domain Criteria approach to psychiatric showed a thicker right rACC (pcorrected = 0.042) and a disease stresses the importance of identifying fundamental thinner right entorhinal cortex (pcorrected = 0.042). symptom dimensions tied to neural systems that cut Although none of six surface area by group interaction across heterogeneous mental disorder classifications37. effects reached significance (p > 0.074), interaction effects Our findings are the first to identify brain structural for the two identified cortical thickness regions did (all correlates related to insomnia, an important clinical p < 0.016), supporting specificity for these regions to symptom of the Arousal and Regulatory Systems healthy controls but not for MDD patients. domain38, in people suffering from MDD. Notably how- Together these results suggest differential association ever, these correlates do not seem to cut across disorders. profiles of cortical measures in MDD that in general do Our findings indicate that only cortical surface area is not generalize to BD clinical controls or healthy controls. predictive of insomnia severity in MDD, whereas cortical thickness and subcortical volume had no predictive value. Discussion Prior studies have shown that these measures represent This large-scale study investigated brain structural corre- distinct biological processes. For example, cortical surface lates of insomnia severity in MDD and revealed more severe area, cortical thickness, and GM volume differ in terms of insomnia in cases with a smaller total surface area. This developmental trajectory39, network topology40, and inverse association with total surface area was mostly driven genetic influences41. As compared to cortical thickness, Leerssen et al. Translational Psychiatry (2020)10:425 Page 8 of 11 surface area is more strongly determined by genetic The cortical regions for which a smaller surface area influences42. To identify common genetic variants that predicted more severe insomnia are involved in a wide underlie these genetic influences on brain structures is range of functions, including emotional processing (medial not straightforward, as their effects are very small. To OFC, frontal pole, insula), attentional processing and overcome this difficulty, >50 ENIGMA sites recently interoceptive awareness (insula), and cognitive control generated a very large sample (n = 35,660) to uncover (inferior frontal gyrus pars triangularis, insula, parietal genetic loci that affect cortical surface area and thick- regions)46,47. It may—at first glance—be surprising that ness42. The study revealed many loci where variants were insomnia severity is significantly associated with the surface associated with surface area. Most interestingly, genetic area of regions that are primarily involved in these pro- correlations indicated that the variants associated with a cesses, while overall depression severity is not. Recent smaller global surface area overlapped more with the insights, however, suggest that insomnia involves altered variants involved in insomnia6 than with variants of any emotion regulation and interoception rather than deficits other included symptom or disorder. In light of (1) the in sleep regulation per se48–50, which is again supported by strong genetic correlation between insomnia and cortical our findings of reduced surface area in regions involved in surface area, (2) the genetic heritability of surface area, emotional processing. and (3) the more externally driven variability of cortical The current study has several limitations. First, we had thickness, we consider it likely that overlapping neuro- limited information on sleep in our sample: only three biological mechanisms predispose to both a smaller cor- HDRS items about insomnia. It would have been inter- tical surface area and more severe insomnia symptoms in esting to evaluate whether cortical surface areas showed MDD. We cannot fully exclude, however, the possibility similar associations with other measures of sleep, as could that insomnia causes a reduction of cortical surface area be derived from sleep diaries, actigraphy, or poly- as secondary process. somnography. Even so, actigraphic and polysomnographic We found smaller surface areas of several cortical regions measures of sleep hardly correlate with the subjective to be associated with insomnia severity in MDD patients; complaints that diagnostically define insomnia51. By such associations were, however, not found in non- contrast, subjective complaints recorded in sleep diaries depressed samples. Few studies investigated cortical sur- strongly correlate with the insomnia items of the HDRS27. face area in relation to insomnia complaints. Lim et al.43 Second, the characteristics of the HCP healthy controls found that sleep fragmentation was nominally associated were somewhat different: they were younger, scanned on with lower surface area in the banks of the superior tem- a different scanner, and asked different insomnia ques- poral sulcus and pars orbitalis. While we did not find tions than in the ENGIMA MDD sample. Nevertheless, cortical thickness to be associated with insomnia severity in these results still provide valuable insight into how MDD, we did find insomnia severity to be associated with insomnia-related brain alterations may be different in thickness alterations in the entorhinal cortex and the rACC people with MDD than in people without MDD. Third, in our healthy control sample. Several studies have repor- poor sleep quality might be associated with obstructive ted an association between thickness and insomnia severity sleep apnea, a late chronotype, and sleep duration. Sleep in non-depressed people43–45. More specifically within apnea and chronotype have been associated with less insomnia patients, one study found thinning in the ACC, GM52 and a thinner cortex53–55; however, as far as we precentral cortex, and lateral prefrontal cortex18, while in know no studies have associated these variables with contrast another study found thickening in several areas, cortical surface area. Insomnia severity might also be including the orbital frontal cortex, rACC, middle cingulate associated with sleep duration; however, in a large study cortex, insula, superior parietal lobule, and fusiform area19. of MDD patients the shared variance between insomnia Concertedly, these findings provide support for a double severity and sleep duration was limited (20%)56, suggest- dissociation suggesting a depression-specific association of ing discernable dimensions of sleep. Unfortunately, sleep insomnia severity with cortical surface area and an asso- apnea, chronotype, and sleep duration were not system- ciation of insomnia severity with cortical thickness in non- atically assessed in our sample. It would be interesting to depressed people. evaluate whether our findings are better explained by Reduced surface area of the medial OFC, however, was these variables than by quality of sleep. Lastly, other found to be related to insomnia severity in both MDD variables could potentially have contributed to individual patients and in healthy controls in our study. One study differences in our dataset, such as handedness57, oral found reduced GM in the medial OFC in co-morbid contraceptive use58, medical comorbidities, or demen- depression and insomnia patients compared to insomnia or tia59–63. Future studies could take these variables into depressed patients without comorbid disorders21. Altera- account. A major strength of our study is that we obtained tions in the medial OFC might have a symptom-specific role data from a large representative sample of MDD patients that is similar in in both insomnia and depressed patients. from 15 different sites, supporting the robustness and Leerssen et al. Translational Psychiatry (2020)10:425 Page 9 of 11 generalizability of our results. The robustness of our of Melbourne & Melbourne Health, Melbourne, VIC, Australia. 4Orygen, The findings is further supported by the lack of interaction National Centre of Excellence in Youth Mental Health, Parkville, VIC, Australia. 5 Imaging Genetics Center, Mark and Mary Stevens Neuroimaging and effects of surface area with antidepressant use, depression Informatics Institute, Keck School of Medicine of USC, University of Southern recurrence, or age of onset of depression. California, Marina del Rey, CA, USA. 6Department of Clinical Neuroscience, In conclusion, our study showed that insomnia is more Behavior & Neurotechnology, Scientific Research Institute of Neuroscience & Medicine, Novosibirsk, Russia. 7Department of Neuroscience, Novosibirsk State severe in patients with MDD who have a smaller cortical University, Novosibirsk, Russia. 8NORMENT Centre, Division of Mental Health surface area, in particular of the right insula, left inferior and Addiction, Oslo University Hospital, Oslo, Norway. 9Institute of Clinical frontal gyrus pars triangularis, left frontal pole, right Medicine, University of Oslo, Oslo, Norway. 10Department of Psychiatry, Melbourne Medical School, University of Melbourne, Parkville, VIC, Australia. superior parietal cortex, right medial OFC, and right 11 Department of Psychiatry, University of Münster, Münster, Germany. supramarginal gyrus. The better specificity of these 12 Department of Psychiatry, Trinity College Dublin, Dublin, Ireland. 13 associations with insomnia severity than with total Department of Psychiatry and Psychotherapy, Otto von Guericke University Magdeburg, Magdeburg, Germany. 14Psychopharmacology Research Unit, depression severity highlights the possibility that insom- Department of Psychiatry, University of Oxford, Oxford, UK. 15Department of nia could represent a symptom cluster of MDD with a Psychology, Stanford University, Stanford, CA, USA. 16Section for Experimental distinct neurobiological underpinning. Psychopathology and Neuroimaging, Department of General Psychiatry, Heidelberg University, Heidelberg, Germany. 17Division of Psychiatry, University of Edinburgh, Edinburgh, UK. 18Youth Mental Health Team, Brain and Mind Acknowledgements Centre, University of Sydney, Sydney, NSW, Australia. 19University of Ottawa’s This project has received funding from the European Research Council (ERC) Institute of Mental Health Research, Ottawa, ON, Canada. 20Cellular & Molecular under the European Union’s Horizon 2020 research and innovation program Medicine, University of Ottawa, Ottawa, ON, Canada. 21Department of (grant agreement No. 671084 - INSOMNIA, J.L., T.F.B., E.J.W.V.S.) and from the VU Psychiatry and Psychotherapy, Philipps-University Marburg, Marburg, Germany. University Amsterdam University Research Fellowships (J.L., T.F.B.). The 22 Department of Psychiatry and Psychotherapy, University of Bonn, Bonn, ENIGMA-MDD Working Group is supported by NIH Consortium grant U54 Germany. 23Sunshine Coast Mind and Neuroscience Thompson Institute, EB020403, R01 MH116147, P41 EB015922 (P.M.T., N.J.) and R01 MH117601 (N.J., Birtinya, QLD, Australia. 24Department of Psychiatry and Psychotherapy, Jena L.S.). This work was further supported by a “Miguel Servet-II” contract from the University Hospital, Jena, Germany. 25Max Planck Institute for Biological Spanish Ministry of Health (CP16-00020) and partly funded by PI13/01057 (M.J. Cybernetics, Tübingen, Germany. 26Psychiatry and Paediatrics, University of P.), Wellcome Trust Grant 10436/Z/14/Z (A.M.M.), and the European Calgary, Calgary, AB, Canada. 27Strategic Clinical Network for Addictions and Community’s Seventh Framework Programme (FP7/2007-2013) under grant Mental Health, Calgary, AB, Canada. 28Centre for Cognitive Ageing and agreement no. 602450. This paper reflects only the author’s views and the Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK. 29Faculty of European Union is not liable for any use that may be made of the information Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada. contained therein. This work was also supported by a Wellcome Trust Strategic 30 Department of Psychiatry, University of Texas Health Science Center at Award 104036/Z/14/Z and the IMAGEMEND grant (H.C.W.); Branch Out Houston, Houston, TX, USA. 31Department of Psychiatry, Institute of Biomedical Neurological Foundation, Children’s Hospital Aid Society (F.P.M.); Science Research Sant Pau, Barcelona, Spain. 32CIBERSAM, Barcelona, Spain. Foundation Ireland (SFI) Professorship grant (T.F.); German Research 33 Department of Psychology, University of Halle, Halle, Germany. 34Center for Foundation (DFG, grant FOR2107 KR 3822/5-1 and KR 3822/7-2 to A.K.; DFG, Depression, Anxiety, and Stress Research, McLean Hospital, Harvard Medical grant FOR2107 KI 588/14-1 and KI 588/14-2 to T.K.; DFG, grant FOR2107 NE School, Belmont, MA, USA. 35Max Planck Institute of Psychiatry, Munich, 2254/1-2 to I.N.; DFG, grant FOR2107 DA1151/5-1 and DA1151/5-2 to U.D.); Germany. 36UT Center of Excellence on Mood Disorders, Department of NIMH grant R01 085667, the Dunn Foundation and the Pat Rutherford, Jr. Psychiatry and Behavioral Sciences, University of Texas Health Science Center Endowed Chair in Psychiatry (J.C.S.); German Research Foundation (SFB-TRR58, at Houston, Houston, TX, USA. 37Department of Psychiatry and Psychotherapy, Projects C09 and Z02 to U.D.) and the Interdisciplinary Center for Clinical University of Tübingen, Tübingen, Germany. 38Department of Health Research (IZKF) of the medical faculty of Münster (grant Dan3/012/17 to U.D.); Promotion and Human Behavior, Graduate School of Public Health/ School of Optimising Personalised Care, at scale, for Young People with Emerging Mood Public Health, Kyoto University, Kyoto, Japan. 39Center for Health Sciences, SRI Disorders $951,005 over 5 years (2018-2022) (APP1136259, I.B.H.). Russian International, Menlo Park, CA, USA. 40Department of Psychiatry, Amsterdam Science Foundation grant #16-15-00128 (L.A.); UK Medical Research Council, UMC, Amsterdam Neuroscience, VU University, Amsterdam, Netherlands. Grant number G0701421, (B.R.G.); NIMH Grants R01-MH59259 and R37- 41 Amsterdam Neuroscience, VU University Medical Center, Amsterdam, MH101495 (I.H.G.); NIA T32AG058507, NIH/NIMH 5T32MH073526, NIH grant Netherlands. 42Centre for Youth Mental Health, The University of Melbourne, U54EB020403 from the Big Data to Knowledge (BD2K) Program (C.R.K.C.); Melbourne, VIC, Australia Wellcome Trust Strategic Award “STratifying Resilience and Depression Longitudinally” (STRADL) Reference 104036/Z/14/Z (E.H.); Dr Mortimer and Theresa Sackler Foundation (M.A.H.); Innovative Medical Research (RE111604 and RE111722 to R.R.); Medical Review, Sogen-sha (N.W.); Alberta Children’s Code availability Hospital Research Institute (Q.M.). MPIP data were in part provided by the Analysis scripts are available upon reasonable request by contacting the Munich Antidepressant Response Signature study, supported by a grant of the corresponding author. Exzellenz-Stiftung of the Max Planck Society and funded by the Federal Ministry of Education and Research (BMBF) in the framework of the National Genome Research Network (NGFN), FKZ 01GS0481. Data were provided (in Conflict of interest part) by the Human Connectome Project, WU-Minn Consortium (Principal I.B.H. has previously led community-based and pharmaceutical industry- Investigators: David Van Essen and Kamil Ugurbil; 1U54MH091657) funded by supported (Wyeth, Eli Lily, Servier, Pfizer, AstraZeneca) projects focused on the the 16 NIH Institutes and Centers that support the NIH Blueprint for identification and better management of anxiety and depression. He was a Neuroscience Research and by the McDonnell Center for Systems member of the Medical Advisory Panel for Medibank Private until October Neuroscience at Washington University. 2017, a Board Member of Psychosis Australia Trust, and a member of Veterans Mental Health Clinical Reference group. He is the Chief Scientific Advisor to, Author details and an equity shareholder in, Innowell. Innowell has been formed by the 1 Department of Sleep and Cognition, Netherlands Institute for Neuroscience, University of Sydney and PwC to deliver the $30 m Australian Government- an Institute of the Royal Netherlands Academy of Arts and Sciences, funded “Project Synergy.” Project Synergy is a 3-year program for the Amsterdam, The Netherlands. 2Department of Integrative Neurophysiology, transformation of mental health services through the use of innovative Center for Neurogenomics and Cognitive Research (CNCR), Amsterdam technologies. C.R.K.C. received partial research support from Biogen, Inc. Neuroscience, VU University Amsterdam, Amsterdam, The Netherlands. (Boston, USA) for work unrelated to the topic of this manuscript. The other 3 Melbourne Neuropsychiatry Centre, Department of Psychiatry, The University authors report no financial relationships with commercial interests. Leerssen et al. Translational Psychiatry (2020)10:425 Page 10 of 11 Publisher’s note 21. Yu, S. et al. The orbitofrontal cortex gray matter is associated with the inter- Springer Nature remains neutral with regard to jurisdictional claims in action between insomnia and depression. Front. Psychiatry 9, 651 (2018). published maps and institutional affiliations. 22. Benjamins, J. S. et al. Insomnia heterogeneity: characteristics to consider for data-driven multivariate subtyping. Sleep. Med. Rev. 36, 71–81 (2017). Supplementary Information accompanies this paper at (https://doi.org/ 23. Blanken, T. F. et al. Insomnia disorder subtypes derived from life history and 10.1038/s41398-020-01109-5). traits of affect and personality. Lancet Psychiatry 6, 151–163 (2019). 24. Hamilton, M. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56–62 (1960). Received: 8 April 2020 Revised: 4 November 2020 Accepted: 17 November 25. Van Essen, D. C. et al. The WU-Minn Human Connectome Project: an overview. 2020 Neuroimage 80, 62–79 (2013). 26. Buysse, D. J., Reynolds, C. F. III, Monk, T. H., Berman, S. R. & Kupfer, D. J. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 28, 193–213 (1989). 27. Manber, R. et al. Assessing insomnia severity in depression: comparison References of depression rating scales and sleep diaries. J. Psychiatr. Res. 39, 1. Benca, R. M., Obermeyer, W. H., Thisted, R. A. & Gillin, J. C. Sleep and psychiatric 481–488 (2005). disorders: a meta-analysis. Arch. Gen. Psychiatry 49, 651–668 (1992). 28. Achenbach T. M. The Achenbach System of Empirically Based Assessment 2. Baglioni, C. et al. Insomnia as a predictor of depression: a meta-analytic eva- (ASEBA): Development, Findings, Theory, and Applications (University of Vermont, luation of longitudinal epidemiological studies. J. Affect. Disord. 135, 10–19 Research Center for Children, Youth, & Families, 2009). (2011). 29. Schmaal, L. et al. Cortical abnormalities in adults and adolescents with 3. Thase, M. E. et al. Which depressed patients will respond to interpersonal major depression based on brain scans from 20 cohorts worldwide in psychotherapy? The role of abnormal EEG sleep profiles. Am. J. Psychiatry 154, the ENIGMA Major Depressive Disorder Working Group. Mol. Psychiatry 502–509 (1997). 22, 900–909 (2017). 4. Cunningham, J. E. A. & Shapiro, C. M. Cognitive behavioural therapy for 30. Schmaal, L. et al. Subcortical brain alterations in major depressive disorder: insomnia (CBT-I) to treat depression: a systematic review. J. Psychosom. Res. findings from the ENIGMA Major Depressive Disorder working group. Mol. 106, 1–12 (2018). Psychiatry 21, 806–812 (2016). 5. Blanken, T. F. et al. Introducing network intervention analysis to investigate 31. Glasser, M. F. et al. The minimal preprocessing pipelines for the Human sequential, symptom-specific treatment effects: a demonstration in co- Connectome Project. Neuroimage 80, 105–124 (2013). occurring insomnia and depression. Psychother. Psychosom. 88, 52–54 32. Fischl, B. et al. Whole brain segmentation: automated labeling of neuroana- (2019). tomical structures in the human brain. Neuron 33, 341–355 (2002). 6. Hammerschlag, A. R. et al. Genome-wide association analysis of insomnia 33. Desikan, R. S. et al. An automated labeling system for subdividing the human complaints identifies risk genes and genetic overlap with psychiatric and cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage metabolic traits. Nat. Genet. 49, 1584–1592 (2017). 31, 968–980 (2006). 7. Jansen, P. R. et al. Genome-wide analysis of insomnia in 1,331,010 individuals 34. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical identifies new risk loci and functional pathways. Nat. Genet. 51, 394–403 and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. (2019). 57, 289–300 (1995). 8. Altena, E., Vrenken, H., Van Der Werf, Y. D., van den Heuvel, O. A. & Van 35. Leon, A. C. & Heo, M. Sample sizes required to detect interactions between Someren, E. J. Reduced orbitofrontal and parietal gray matter in chronic two binary fixed-effects in a mixed-effects linear regression model. Comput. insomnia: a voxel-based morphometric study. Biol. Psychiatry 67, 182–185 Stat. Data Anal. 53, 603–608 (2009). (2010). 36. Button, K. S. et al. Power failure: why small sample size undermines the 9. Winkelman, J. W. et al. Increased rostral anterior cingulate cortex volume in reliability of neuroscience. Nat. Rev. Neurosci. 14, 365 (2013). chronic primary insomnia. Sleep 36, 991–998 (2013). 37. Insel, T. et al. Research Domain Criteria (RDoC): Toward a New Classification 10. Joo, E. Y. et al. Brain gray matter deficits in patients with chronic primary Framework for Research on Mental Disorders (American Psychiatric Association, insomnia. Sleep 36, 999–1007 (2013). 2010). 11. Li, M. et al. Altered gray matter volume in primary insomnia patients: a 38. Cuthbert, B. N. & Insel, T. R. Toward the future of psychiatric diagnosis: the DARTEL-VBM study. Brain Imaging Behav. 12, 1759–1767 (2018). seven pillars of RDoC. BMC Med. 11, 126 (2013). 12. Bumb, J. M. et al. Pineal gland volume in primary insomnia and healthy 39. Wierenga, L. M., Langen, M., Oranje, B. & Durston, S. Unique developmental controls: a magnetic resonance imaging study. J. Sleep. Res. 23, 274–280. trajectories of cortical thickness and surface area. Neuroimage 87, 120–126 (2014). (2014). 13. Li, M. et al. Abnormalities of thalamus volume and resting state functional 40. Sanabria-Diaz, G. et al. Surface area and cortical thickness descriptors reveal connectivity in primary insomnia patients. Brain Imaging Behav. 13, 1193–1201 different attributes of the structural human brain networks. Neuroimage 50, (2019). 1497–1510 (2010). 14. Kang, J. M. et al. Low white-matter integrity between the left thalamus and 41. Winkler, A. M. et al. Cortical thickness or grey matter volume? The importance inferior frontal gyrus in patients with insomnia disorder. J. Psychiatry Neurosci. of selecting the phenotype for imaging genetics studies. Neuroimage 53, 43, 170195 (2018). 1135–1146 (2010). 15. Riemann, D. et al. Chronic insomnia and MRI-measured hippocampal 42. Grasby, K. L. et al. The genetic architecture of the human cerebral cortex. volumes: a pilot study. Sleep 30, 955–958 (2007). Science 367, eaay6690 (2020). 16. Joo, E. Y., Kim, H., Suh, S. & Hong, S. B. Hippocampal substructural vulnerability 43. Lim, A. S. P. et al. Regional neocortical gray matter structure and sleep frag- to sleep disturbance and cognitive impairment in patients with chronic pri- mentation in older adults. Sleep 39, 227–235 (2016). mary insomnia: magnetic resonance imaging morphometry. Sleep 37, 44. Kocevska, D. et al. The developmental course of sleep disturbances across 1189–1198 (2014). childhood relates to brain morphology at age 7: the Generation F study. Sleep 17. Leerssen, J. et al. Increased hippocampal-prefrontal functional connectivity in 40, zsw022 (2017). insomnia. Neurobiol. Learn. Mem. 160, 144–150 (2019). 45. Sexton, C. E., Storsve, A. B., Walhovd, K. B., Johansen-Berg, H. & Fjell, A. M. Poor 18. Suh, S., Kim, H., Dang-Vu, T. T., Joo, E. & Shin, C. Cortical thinning and altered sleep quality is associated with increased cortical atrophy in community- cortico-cortical structural covariance of the default mode network in patients dwelling adults. Neurol 83, 967–973 (2014). with persistent insomnia symptoms. Sleep 39, 161–171 (2016). 46. Damasio, A. R. et al. Subcortical and cortical brain activity during the feeling of 19. Yu, S. et al. Gray matter hypertrophy in primary insomnia: a surface-based self-generated emotions. Nat. Neurosci. 3, 1049 (2000). morphometric study. Brain Imaging Behav. 14, 1309–1317 (2020). 47. Menon, V. & Uddin, L. Q. Saliency, switching, attention and control: a network 20. Vassilopoulou, K. et al. A magnetic resonance imaging study of hippocampal, model of insula function. Brain Struct. Funct. 214, 655–667 (2010). amygdala and subgenual prefrontal cortex volumes in major depression 48. Wassing, R. et al. Slow dissolving of emotional distress contributes to hyper- subtypes: melancholic versus psychotic depression. J. Affect. Disord. 146, arousal. Proc. Natl Acad. Sci. USA 113, 2538–2543 (2016). 197–204 (2013). Leerssen et al. Translational Psychiatry (2020)10:425 Page 11 of 11 49. Wei, Y. et al. I keep a close watch on this heart of mine: increased inter- depressive disorder and anxiety disorders. J. Clin. Psychiatry 71, 239–246 oception in insomnia. Sleep 39, 2113–2124 (2016). (2010). 50. Wassing, R., Benjamins, J. S., Talamini, L. M., Schalkwijk, F. & Van Someren, E. J. 57. Willems, R. M., der Haegen, L. V., Fisher, S. E. & Francks, C. On the other hand: W. Overnight worsening of emotional distress indicates maladaptive sleep in including left-handers in cognitive neuroscience and neurogenetics. Nat. Rev. insomnia. Sleep 42, zsy268 (2019). Neurosci. 15, 193–201 (2014). 51. Rosa, R. R. & Bonnet, M. H. Reported chronic insomnia is independent of poor 58. Petersen, N., Touroutoglou, A., Andreano, J. M. & Cahill, L. Oral contraceptive sleep as measured by electroencephalography. Psychosom. Med. 62, 474–482 pill use is associated with localized decreases in cortical thickness. Hum. Brain (2000). Mapp. 36, 2644–2654 (2015). 52. Shi, Y. et al. A meta-analysis of voxel-based brain morphometry studies in 59. Storsve, A. B. et al. Differential longitudinal changes in cortical thickness, sur- obstructive sleep apnea. Sci. Rep. 7, 10095 (2017). face area and volume across the adult life span: regions of accelerating and 53. Joo, E. Y., Jeon, S., Kim, S. T., Lee, J.-M. & Hong, S. B. Localized cortical thinning decelerating change. J. Neurosci. 34, 8488–8498 (2014). in patients with obstructive sleep apnea syndrome. Sleep 36, 1153–1162 60. Dotson, V. M. et al. Age differences in prefrontal surface area and (2013). thickness in middle aged to older adults. Front. Aging Neurosci. 7, 250 54. Macey, P. M. et al. Obstructive sleep apnea and cortical thickness in females (2016). and males. PLoS ONE 13, e0193854 (2018). 61. Sepulcre, J. et al. Neurogenetic contributions to amyloid beta and tau 55. Rosenberg, J., Jacobs, H. I. L., Maximov, I. I., Reske, M. & Shah, N. J. Chronotype spreading in the human cortex. Nat. Med. 24, 1910–1918 (2018). differences in cortical thickness: grey matter reflects when you go to bed. 62. Sepulcre, J. et al. Hierarchical organization of tau and amyloid deposits in the Brain Struct. Funct. 223, 3411–3421 (2018). cerebral cortex. JAMA Neurol. 74, 813–820 (2017). 56. van Mill, J. G., Hoogendijk, W. J., Vogelzangs, N., van Dyck, R. & Penninx, B. W. 63. Cho, H. et al. Predicted sequence of cortical tau and amyloid-β deposition in Insomnia and sleep duration in a large cohort of patients with major Alzheimer disease spectrum. Neurobiol. Aging 68, 76–84 (2018).