Skip to main content
NCBI home page
Sleep logoLink to Sleep
. 2013 May 1;36(5):731–737. doi: 10.5665/sleep.2638

Insomnia Does Not Appear to be Associated With Substantial Structural Brain Changes

Kai Spiegelhalder 1,✉, Wolfram Regen 1, Chiara Baglioni 1, Stefan Klöppel 1, Ahmed Abdulkadir 1, Jürgen Hennig 2, Christoph Nissen 1, Dieter Riemann 1, Bernd Feige 1
PMCID: PMC3624828  PMID: 23633756

Abstract

Study Objectives:

Sleep has been demonstrated to significantly modulate brain plasticity and the manifestation of mental disorders. However, previous studies on the effect of disrupted sleep on brain structure have reported inconsistent results. The goal of the current study was to investigate brain morphometry in a well-characterized large sample of patients with primary insomnia (PI) in comparison with good sleeper controls.

Design:

Automated parcellation and pattern recognition approaches were supplemented by voxelwise analyses of gray and white matter volumes to analyze magnetic resonance images. All analyses included age, sex, and total intracranial volume as covariates.

Setting:

Department of Psychiatry and Psychotherapy of the University of Freiburg Medical Center.

Participants:

There were 28 patients with PI (10 males; 18 females; age 43.7 ± 14.2 y) and 38 healthy, good sleepers (17 males; 21 females; age 39.6 ± 8.9 y).

Interventions:

N/A.

Results:

No significant between-group differences were observed in any of the investigated brain morphometry variables.

Conclusions:

Altered brain function in insomnia does not appear to have a substantial effect on brain morphometry on a macroscopic level.

Citation:

Spiegelhalder K; Regen W; Baglioni C; Klöppel S; Abdulkadir A; Hennig J; Nissen C; Riemann D; Feige B. Insomnia does not appear to be associated with substantial structural brain changes. SLEEP 2013;36(5):731-737.

Keywords: Hippocampus, MRI, primary insomnia, sleep, support vector machine, voxel-based morphometry

INTRODUCTION

Chronic insomnia is one of the most prevalent health complaints worldwide and its prevalence is expected to further increase with the aging of society.1 It affects approximately 10% of the population in western industrialized countries.2 Primary insomnia (PI), defined as sleep complaints in the absence of any related medical or psychiatric condition, is estimated to affect 2% to 4% of the adult population.2 In most affected individuals, insomnia is a chronic condition. More than 70% of the people currently experiencing insomnia will still suffer from it next year.3 Chronic insomnia is associated with a marked reduction in quality of life, increased fatigue, cognitive impairments, mood disturbances, and physical complaints.4,5 Importantly, insomnia confers an increased risk for psychiatric disorders, especially depression,6 and there is evidence that chronic sleep loss is a risk factor for cardiovascular disease and increased mortality.7 Accordingly, insomnia leads to a substantial increase in health care consumption and to a high rate of absenteeism, and consequently, to high costs for society. For the United States, the costs of insomnia due to low work performance and absenteeism have been estimated to exceed $60 billion per year.8 Thus, insomnia significantly contributes to the major diseases of our aging society and to a significant part of the health expenses.

Central nervous system hyperarousal is believed to represent a major pathophysiologic pathway in the development and maintenance of insomnia9–11; however, few neuroimaging studies have examined the neurologic abnormalities of the disorder. Three recent investigations have focused on the association between insomnia and morphometric changes of the brain.12–14 In a pilot study, we investigated eight patients with PI and eight healthy good sleepers with a manual tracing method and found reduced bilateral hippocampal volumes,12 which was supported by basic science research showing that chronic sleep deprivation has an effect on hippocampal morphology.15,16 Furthermore, sleep duration was shown to be associated with hippocampal volumes in children17 and insomnia severity was associated with hippocampal volumes in a study on patients with posttraumatic stress disorder.18 However, a study by Winkelman et al.13 failed to show reduced hippocampus gray matter volumes using manual tracing in 20 patients with PI in comparison with 15 healthy good sleepers. These inconsistent results might stem from methodologic differences (e.g., different segmentation protocols) and heterogeneity with regard to insomnia duration and severity. Additionally, age differences between study samples might account for differing results. Winkelman et al.13 investigated a community sample with a mean age of approximately 40 y and used a segmentation protocol that was restricted to core areas of the hippocampus. We, instead, investigated clinically referred patients with a mean age of approximately 48 y and followed the Watson protocol to measure the volumes of the hippocampus.19 In the third morphometric study in insomnia patients, Altena et al.14 used voxel-based morphometry without a priori definition of regions of interest investigating an elderly sample (24 patients with PI, 13 healthy good sleepers) with a mean age of approximately 60 y. In this sample, the authors reported reduced orbitofrontal and precuneus gray matter volumes but no between-group differences in the hippocampal region.

In summary, previous findings regarding brain morphology alterations of insomnia patients have been based on relatively small samples and provided inconsistent results. Therefore, the current study aimed at investigating brain morphometry in a clinically referred and well-characterized large sample of patients with PI. We extended previous work by using a multi-method approach including an automated brain segmentation algorithm (FreeSurfer software, version 5.1.0 (http://surfer.nmr.mgh.harvard.edu/),20 an automated classification algorithm (support vector machine)21 based on FreeSurfer-derived cortical volume and thickness estimates, and voxel-based morphometry.22 In comparison with the other methods, the machine-learning classification algorithm is multivariate and takes specific interregional dependencies into account. Its high sensitivity has been demonstrated for a number of neurologic and psychiatric conditions.23 The specific hypotheses of the current study were: (1) patients with PI have reduced gray matter volumes in automatically segmented hippocampal volumes; (2) magnetic resonance imaging (MRI)-derived brain volume and cortical thickness estimates can reliably distinguish patients with PI from healthy good sleepers using an automated classification algorithm; (3) voxel-based morphometry reveals gray and/or white matter volume differences between patients with PI and healthy good sleepers.

MATERIAL AND METHODS

Participants

Twenty-eight patients meeting diagnostic criteria for PI according to the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision24 and 40 good sleeper controls were included in the current study. Insomnia patients were referred to our sleep disorders clinic by their primary care provider or medical specialist. Healthy controls were recruited through local advertisements. Two control participants were excluded from the analysis because of pathologic MRI scans. Thus, the final sample consisted of 28 patients with PI and 38 good sleeper controls.

A semistandardized psychiatric and sleep-related interview was conducted by an experienced psychiatrist to rule out any history of psychiatric disorder, shift work, or sleep disorder (including hypersomnia, parasomnia, sleep related breathing disorder, sleep related movement disorder, and circadian rhythm sleep disorder). Furthermore, all participants underwent a standard physical examination, including electrocardiogram, electroencephalogram (EEG), and routine blood work (blood cell count; liver, renal and thyroid function) to exclude those with serious medical conditions. All participants were right-handed, as assessed with the Edinburgh Handedness Inventory25 and free of any psychoactive medication at least 2 weeks prior to and during the study. Participants with periodic leg movements during sleep arousal index per total sleep time (TST) of more than 5.0 or a sleep apnea index per TST of more than 5.0 were not included in the current study.

The study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of the University Medical Center Freiburg. All participants gave their informed written consent prior to inclusion in the study.

Polysomnography

All participants underwent 2 consecutive nights of polysomnography sleep monitoring. The first night served as an adaptation and screening night to rule out sleep apnea, periodic leg movements in sleep, and occult sleep disorder pathology. Sleep was recorded on 24-channel Sagura EEG-polysomnographs for 8 h from “lights out” (22:00 to 23:00) until “lights on” (06:00 to 07:00). All recordings included EEG (C3-A2; C4-A1), electrooculogram (horizontal and vertical) and electromyogram (submental), and were scored visually by experienced raters according to the American Academy of Sleep Medicine criteria.26 All participants were screened for apneas and periodic leg movements by monitoring abdominal and thoracic effort, nasal airflow, oxymetry, and bilateral tibialis anterior EMG. Sleep recordings were evaluated for the following parameters of sleep continuity: TST; sleep efficiency (ratio of TST to time in bed × 100%); sleep onset latency defined as time from lights out until sleep onset (defined as first epoch of stage 2); wake after sleep onset (WASO) defined as difference between sleep period time (SPT; time from sleep onset until final awakening) and TST; number of awakenings; and arousal index. Sleep architecture parameters were amounts of stages 1 and 2 slow wave sleep (SWS) and rapid eye movement sleep (REM) as percentage of SPT. All participants had to refrain from alcohol, caffeine, and daytime naps during the recording days.

Questionnaires

Participants were asked to complete the Insomnia Severity Index (ISI),27 the Pittsburgh Sleep Quality Index (PSQI),28 the brief version of the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS-16),29 the Glasgow Sleep Effort Scale (GSES),30 the Pre-Sleep Arousal Scale (PSAS),31 the Ep-worth Sleepiness Scale (ESS),32 the Beck Depression Inventory (BDI),33 and the State-Trait Anxiety Inventory (STAI).34

MRI Acquisition and Analysis

High-resolution T1-weighted MRI datasets were acquired on a 3-Tesla scanner (Magnetom TIM-Trio, Siemens, Erlangen, Germany) using an MPRAGE sequence (TR 2.2 sec; TE 2.6 msec; 160 sagittal slices of 256 × 256 voxels, 1.0 × 1.0 × 1.0 mm3).35 All scans were inspected for motion artefacts and for the absence of pathologic findings by a neurologist under the supervision of a board-certified neuroradiologist.

FreeSurfer-Based Univariate Analyses

Cortical surface reconstruction and volumetric segmentation was performed using the FreeSurfer software, version 5.1.0 (Athinoula A. Martinos Center for Biomedical Imaging; http://surfer.nmr.mgh.harvard.edu/). The technical details of this procedure have been described in previous publications20,36,37 and good test-retest reliability has been demonstrated.38 For quality control, segmentations were visually inspected for each participant by two independent raters (K.S. and W.R.) on a slice-by-slice basis. However, no manual corrections were necessary for the automatic segmentation results. Subcortical volumes (“aseg.stats” files) as well as cortical volume and thickness values (“aparc.stats” files) based on the Killiany/Desikan cortical parcellation39 were extracted from FreeSurfer, resulting in a total of 174 measures per participant. Statistical analyses were conducted in R (http://www.r-project.org/) with intracranial volume (ICV), age, and sex as covariates in all analyses. Analyses of covariance (ANCOVA) were carried out for investigating between-group differences in left and right hippocampal volumes. In these analyses, an uncorrected statistical threshold of P < 0.05 was used due to the a priori hypothesis concerning the hippocampus. Because several ANCOVAs were performed to assess between-group differences in hippocampus subsegments (subiculum, presubiculum, CA1, CA2/3, CA4/dentate gyrus, hippocampal fissure, and fimbria, as derived by the hippocampal subfield segmentation procedure of FreeSurfer), the thresholds for significance (P < 0.05) were adjusted to control for the false discovery rate (FDR).40 Additionally, regression analyses were conducted to analyze the association between hippocampal volumes and the following parameters: ISI scores, TST in the second night, duration of insomnia, BDI scores, and trait STAI scores. The effect of lifetime history of hypnotic medication (none, benzodiazepine receptor agonists, other) on hippocampal volumes was determined by an ANCOVA analysis in the patients group. Between-group differences in the other FreeSurfer-derived measures were also tested with an FDR-adjusted threshold of P < 0.05 due to multiple testing. Analogous to the previously described analyses, between-group differences in hippocampal volumes and in the other FreeSurfer-derived variables were additionally investigated for the subsamples of sleep maintenance insomnia patients and mixed insomnia patients in comparison with the control group.

Support Vector Machine

A support vector machine41 (SVM) was implemented in libsvm, an integrated software for support vector classification (http://www.csie.ntu.edu.tw/∼cjlin/libsvm/). The SVM is a supervised pattern recognition approach. It is supervised in the sense that it learns to separate diagnostic groups (i.e., patients with PI versus good sleeper controls) from example images. The pattern detected from training data is then applied to new data not involved in the training. The SVM has two open parameters, the cost factor, C, and the filter fraction, F. The cost factor determines how much a misclassified example is penalized during training. The second parameter is the fraction of the features, ranked according to their t-score, that are used for classification. We tested the performance with either all, 1/2, 1/4, 1/8, or 1/16 of the features included in the analyses. The two parameters were optimized in a nested (two-level) leave-one-example-out cross-validation (LOO-CV) procedure to avoid optimistic results. Given N training examples, the LOOCV procedure repeatedly leaves one example out, trains an “optimal model” with the parameter combination that yielded the highest performance using N-1 remaining examples, and then predicts the left-out example. As we included different numbers of patients with PI and good sleeper controls we report the balanced accuracy (BA) as the average of sensitivity and specificity.42 One hundred seventy-four FreeSurfer-derived brain volume and cortical thickness values entered the classification pipeline as attributes describing macroscopic brain anatomy. Age, sex, and ICV entered the classification pipeline as additional attributes. The values were converted into z-scores (i.e., subtraction of mean and division by standard deviation) in order to homogenize the relative a priori relevance for classification. The performance of the classifier was reported as the mean BA, the 95% confidence interval limits, and the P value associated with the hypothesis that an equal or better result was obtained by chance.42

Voxel-Based Morphometry

Voxel-based morphometry (VBM) was performed using the Statistical Parametric Mapping (SPM8) software (Well-come Trust Centre for Neuroimaging; http://www.fil.ion.ucl.ac.uk/spm/software/spm8). Structural images were segmented into gray matter, white matter, and cerebrospinal fluid using the standard segmentation procedure in SPM8, and the results were checked by visual inspection. A gray matter population template with a 1.5-mm cubic resolution was generated using DARTEL (diffeomorphic anatomical registration through exponentiated Lie algebra).43 DARTEL-registered data were affine-transformed to Montreal Neurological Institute space (MNI; http://www.mni.mcgill.ca/). Then, all images were modulated to correct for volume changes during normalization, and smoothed with an 8-mm full-width at half-maximum gaussian smoothing kernel. Between-group differences in gray and white matter volumes were assessed using a threshold of P < 0.05, corrected for multiple comparisons by the family-wise error (FWE) method. As Altena et al.14 reported reduced orbitofrontal and parietal gray matter volumes in patients with PI with a more liberal statistical approach, we additionally performed the previously mentioned analyses using an uncorrected voxel-wise threshold of P < 0.001 with a minimum cluster size of 25 voxels. Furthermore, regression analyses were performed to investigate associations between gray/white matter volumes and ISI scores (in the full sample) and insomnia duration (in the PI group) using an FWE-corrected threshold of P < 0.05. Age, sex, and total ICV were included as covariates of no interest in the statistical models applied.

RESULTS

Sample Characteristics

Demographic characteristics of the study sample are presented in Table 1. The groups did not differ significantly in sex distribution, age (patients with PI: 18-67 y, healthy good sleepers: 27-57 y) or body mass index. The patients with PI had significantly higher ISI, PSQI, DBAS-16, and GSES scores as well as higher scores on the cognitive subscale of the PSAS. BDI scores were increased in patients with PI even after excluding the two sleep related items (PI patients: 4.9 ± 3.8, healthy good sleepers: 3.0 ± 2.9, t64 = 2.30, P = 0.025). Furthermore, trait anxiety was increased as indicated by the trait subscale of the STAI. No significant group differences were found for the somatic subscale of the PSAS, the ESS, and the state subscale of the STAI.

Table 1.

Description of the study population (means ± standard deviations)

graphic file with name aasm.36.5.731.t01.jpg

Open in a new tab

Three insomnia patients suffered from sleep onset insomnia, eight from sleep-maintenance insomnia, and 16 from mixed insomnia. One patient had a complaint of nonrestorative sleep in the absence of difficulties initiating and maintaining sleep. The average duration of primary insomnia was 12.1 ± 11.0 y.

Polysomnography

Polysomnographic data are presented in Table 2. Patients with PI had a significantly lower TST, sleep efficiency, and REM% compared with good sleeper controls. Additionally, sleep onset latency in the first night and WASO in both nights were significantly increased in the PI patient group.

Table 2.

Polysomnographic data (means ± standard deviations)

graphic file with name aasm.36.5.731.t02.jpg

Open in a new tab

FreeSurfer-Based Univariate Analyses

ANCOVA analyses revealed no significant between-group differences in left (F62 = 0.11, P = 0.74) and right (F62 < 0.01, P = 0.95) hippocampal volumes (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Left and right hippocampal volumes for patients with primary insomnia (PI) and healthy, good sleepers. Presented volumes are ICV-, age-, and sex-adjusted by using the residual method. Horizontal bars represent group means. GSC, good sleeper controls.

Furthermore, hippocampal subsegments were not statistically different between groups (FDR-corrected). Given the current sample size, the observed variance in hippocampal volumes and an α of 0.05, statistical power analysis revealed that 6.3% (left) and 6.0% (right) between-group differences could have been detected with a power of 80%.

ICV-, age-, and sex-adjusted regression analyses did not reveal any significant association between ISI values and left (t62 = -1.30, P = 0.20) and right (t62 = -0.99, P = 0.33) hippocampal volumes. Furthermore, the TST of the second night in the sleep laboratory was not significantly associated with left (t62 = 0.47, P = 0.64) and right (t62 = -0.68, P = 0.50) hippocampal volumes. With respect to the duration of PI, there were trends for the association with left (t24 = -1.95, P = 0.063, β = -13.0) and right (t24 = -2.00, P = 0.057, β = -12.1) hippo-campal volumes within the PI patient group. However, these trends were not found for the subgroups of sleep-maintenance insomnia patients (left: t4 = -1.44, P = 0.25; right: t4 = -0.58, P = 0.60) or mixed insomnia patients (left: t12 = -0.24, P = 0.82; right: t12 = 0.04, P = 0.97). In the full sample, hippocampal volumes were neither significantly associated with BDI scores (left: t62 = 0.58, P = 0.57; right: t62 = 0.11, P = 0.91) nor with trait STAI scores (left: t62 = 0.91, P = 0.37; right: t62 = 0.94, P = 0.35). Furthermore, lifetime history of hypnotic medication (none, benzodiazepine receptor agonists, other) did not explain differences in the left (F62 = 0.41, P = 0.70) and right (F62 = 1.52, P = 0.24) hippocampal volumes within the PI group.

In the other FreeSurfer-derived brain volume and thickness values, no statistically significant between-group differences were found (FDR-corrected). Uncorrected P values of the ANCOVA analyses are presented in Figure 2, demonstrating that the distribution of P values approximates a random distribution.

Figure 2.

Figure 2

Open in a new tab

Distribution of uncorrected p-values of the between-group comparisons determined by analyses of covariance of 172 FreeSurfer-derived variables (excluding the left and right hippocampus). The slope represents the expected distribution of P values in the absence of significant between-group differences.

The subgroup analyses did not reveal any significant differences to healthy controls in left (sleep maintenance insomnia patients: F42 = 0.08, P = 0.78; mixed insomnia patients: F50 = 0.01, P = 0.91) and right (sleep maintenance insomnia patients: F42 = 0.33, P = 0.57; mixed insomnia patients: F50 = 0.14, P = 0.71) hippocampal volumes. Likewise, in the other Free-Surfer derived variables, no statistically significant between-group differences were found (FDR-corrected).

Support Vector Machine

The cross-validation BA (lower CI-upper CI, P value) using the Freesurfer data was 55.95% (44.32%-67.47%, P = 0.16). Therefore, the hypothesis that the observed performance was due to chance could not be rejected.

Voxel-Based Morphometry

Voxel-based morphometry did not reveal any significant between-group differences in gray or white matter volumes at P < 0.05 (FWE-corrected). Even when using a more liberal statistical approach (uncorrected voxel threshold of P < 0.001 with a cluster threshold of 25 voxels), we did not find any signifi-cant gray or white matter volume differences between patients with PI and good sleeper controls. Furthermore, ICV-, age-, and sex-controlled regression analyses did not reveal any signifi-cant associations between gray and white matter volumes and ISI scores in the full sample, or between gray and white matter volumes and insomnia duration in the PI group (FWE-corrected at P < 0.05).

DISCUSSION

The results of the current study suggest that PI, i.e. insomnia in the absence of a somatic or mental illness, is not associated with substantial macroscopic alterations of the brain structure. In the largest study so far, we used complementary approaches to investigate a clinically referred and well-characterized sample of patients with PI and good sleeper controls. Strengths of the current study include a detailed polysomnographic and psychometric characterization of the sample, the large sample size compared with that of prior studies, and the use of different methods of investigating brain morphometry.

There is overwhelming evidence from psychologic,10,44–47 electrophysiologic,48 and pioneering neuroimaging studies49,50 that brain function is altered in patients with insomnia. However, the current findings suggest that these changes do not relate to substantial alterations of brain morphometry on a macroscopic level. What are the clinical implications for patients with insomnia? They can be reassured that insomnia, according to the current level of knowledge, does not appear to be associated with substantial structural brain changes. This is potentially important because excessive worry about the consequences of poor sleep plays a central role for the development and maintenance of insomnia by triggering emotional and physiologic arousal.10,47

Our power analysis suggests that the absence of statistically significant differences between patients with PI and healthy good sleepers is not due to a lack of statistical power. With respect to hippocampal volumes, between-group differences of approximately 6% could have been detected with a power of 80%. This is in the range of hippocampal volume reductions in other psychiatric disorders. Meta-analyses have revealed bilateral reductions of 5% to 8% in depression,51 posttraumatic stress disorder,52 and schizophrenia.53 However, it has to be noted that we did not have the power to detect smaller brain morphometry changes, which thus can not be ruled out with certainty. The SVM constructs a discriminative model without making any assumptions about the distributions underlying the multivariate data. Thus, it is difficult to convert the sensitivity of an SVM into a power analysis but its high sensitivity has been shown empirically.54–56

For the association between hippocampal volumes and insomnia duration, we observed a statistical trend within the PI patient group. This finding suggests that 10 y of insomnia may be associated with a unilateral volume reduction of 120-130 mm3. However, as the corresponding statistical analyses failed to reach significance, conclusions about this issue have to be drawn carefully. We would like to suggest that future studies on brain morphometry in insomnia may focus specifically on a group of long-term (maybe idiopathic) insomnia patients.

Several limitations of the current investigation have to be acknowledged. First, the patients with PI in the current study were selected without reference to polysomnographically determined sleep parameters. Although this finding is in line with diagnostic criteria and improves the ecologic validity of our results, some studies suggest that polysomnographically determined short sleep duration may be a marker of biologic severity of insomnia.57,58 Accordingly, one may speculate that brain morphometry is specifically altered in the population of insomnia patients with short sleep duration. However, we did not find any significant relationship between insomnia severity as determined by ISI scores or polysomnographically determined TST, and hippocampal volumes in the current study.

Second, patients with PI had higher depression and trait anxiety scores than good sleeper controls. Although none of our participants had a clinically significant affective or anxiety disorder, a subtle effect of subclinical depression or anxiety on brain morphometry cannot be excluded. However, we did not find any significant association between BDI scores or trait STAI scores and hippocampal volumes in the current investigation.

Third, despite a careful screening for relevant sleep problems, we cannot rule out that some of our healthy participants had a period in their life characterized by erratic or chronically restricted sleep due to other reasons than shift work possibly biasing our results. Furthermore, the current sample of healthy controls tended to show comparably poor sleep in contrast to some previous investigations, particularly with respect to the number of awakenings. However, these data are in line with previously published data from a large dataset from our sleep laboratory showing no significant effect of PI on the number of nocturnal awakenings.59 Additionally, a mean sleep efficiency below 90% is not an uncommon finding in polysomnographic investigations of middle-aged healthy samples.60

Fourth, because this was a cross-sectional study with negative findings, conclusions about the significance of the results for insomnia research have to be drawn cautiously. Despite being expensive and very labor intensive, a longitudinal design would allow for more confidence in the conclusions.

In summary, we did not observe any statistically significant brain morphometry changes in a large and well-characterized sample of patients with PI in comparison with good sleeper controls. This might be reassuring for insomnia patients. Yet, the substantial effect of insomnia on quality of life and health care costs is, of course, not affected by a lack of macroscopic brain changes. Given the high prevalence and consequences of insomnia, it remains a public health priority to understand the neurobiology of the disorder to facilitate the development of widely applicable and effective treatment strategies for 10% of the population. Future studies may focus on functional neuro-imaging research in insomnia patients, or, in order to complement the current structural approach, on white matter integrity using diffusion tensor imaging.

DISCLOSURE STATEMENT

This was not an industry supported study. Dr. Nissen has received speaker honoraria from Servier and Novartis. Dr. Riemann has received speaker honoraria from Lundbeck. The other authors have indicated no financial conflicts of interest. Dr. Spiegelhalder and Mr. Regen contributed equally to this work.

ACKNOWLEDGMENTS

Dr. Baglioni and Prof. Dr. Riemann have received funding from the European Community's Seventh Framework Programme (People, Marie Curie Actions, Intra-European Fellowship, FP7-PEOPLE-IEF-2008) under grant agreement number 235321 for this work.

REFERENCES

  • 1.Morin CM, Benca R. Chronic insomnia. Lancet. 2012;379:1129–41. doi: 10.1016/S0140-6736(11)60750-2. [DOI] [PubMed] [Google Scholar]
  • 2.Ohayon M. Epidemiology of insomnia: what we know and what we still need to learn. Sleep Med Rev. 2002;6:97–111. doi: 10.1053/smrv.2002.0186. [DOI] [PubMed] [Google Scholar]
  • 3.Morin CM, Bélanger L, LeBlanc M, et al. The natural history of insomnia: a population-based 3-year longitudinal study. Arch Intern Med. 2009;169:447–53. doi: 10.1001/archinternmed.2008.610. [DOI] [PubMed] [Google Scholar]
  • 4.Kyle SD, Morgan K, Espie CA. Insomnia and health-related quality of life. Sleep Med Rev. 2010;14:69–82. doi: 10.1016/j.smrv.2009.07.004. [DOI] [PubMed] [Google Scholar]
  • 5.NIH. National Institutes of Health state of the science conference statement on manifestations and management of chronic insomnia in adults, June 13-15, 2005. Sleep. 2005;28:1049–57. doi: 10.1093/sleep/28.9.1049. [DOI] [PubMed] [Google Scholar]
  • 6.Baglioni C, Battagliese G, Feige B, et al. Insomnia as a predictor of depression: a meta-analytic evaluation of longitudinal epidemiological studies. J Affect Disord. 2011;135:10–9. doi: 10.1016/j.jad.2011.01.011. [DOI] [PubMed] [Google Scholar]
  • 7.Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiat. 2002;59:131–6. doi: 10.1001/archpsyc.59.2.131. [DOI] [PubMed] [Google Scholar]
  • 8.Kessler RC, Berglund PA, Coulouvrat C, et al. Insomnia and the performance of US workers: results from the American insomnia survey. Sleep. 2011;34:1161–71. doi: 10.5665/SLEEP.1230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Riemann D, Spiegelhalder K, Feige B, et al. The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev. 2010;14:19–31. doi: 10.1016/j.smrv.2009.04.002. [DOI] [PubMed] [Google Scholar]
  • 10.Espie CA, Broomfield NM, MacMahon KM, Macphee LM, Taylor LM. The attention-intention-effort pathway in the development of psychophysiologic insomnia: a theoretical review. Sleep Med Rev. 2006;10:215–45. doi: 10.1016/j.smrv.2006.03.002. [DOI] [PubMed] [Google Scholar]
  • 11.Perlis ML, Giles DE, Mendelson WB, Bootzin RR, Wyatt JK. Psycho-physiological insomnia: the behavioural model and a neurocognitive perspective. J Sleep Res. 1997;6:179–88. doi: 10.1046/j.1365-2869.1997.00045.x. [DOI] [PubMed] [Google Scholar]
  • 12.Riemann D, Voderholzer U, Spiegelhalder K, et al. Chronic insomnia and MRI-measured hippocampal volumes: a pilot study. Sleep. 2007;30:955–8. doi: 10.1093/sleep/30.8.955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Winkelman JW, Benson KL, Buxton OM, et al. Lack of hippocampal volume differences in primary insomnia and good sleeper controls: an MRI volumetric study at 3 Tesla. Sleep Med. 2010;11:576–82. doi: 10.1016/j.sleep.2010.03.009. [DOI] [PubMed] [Google Scholar]
  • 14.Altena E, Vrenken H, van der Werf YD, van den Heuvel OA, van Someren EJ. Reduced orbitofrontal and parietal gray matter in chronic insomnia: a voxel-based morphometric study. Biol Psychiat. 2010;67:182–5. doi: 10.1016/j.biopsych.2009.08.003. [DOI] [PubMed] [Google Scholar]
  • 15.Novati A, Hulshof HJ, Koolhaas JM, Lucassen PJ, Meerlo P. Chronic sleep restriction causes a decrease in hippocampal volume in adolescent rats, which is not explained by changes in glucocorticoid levels or neuro-genesis. Neuroscience. 2011;190:145–55. doi: 10.1016/j.neuroscience.2011.06.027. [DOI] [PubMed] [Google Scholar]
  • 16.Guzman-Marin R, Suntsova N, Methippara M, Greiffenstein R, Syzmusiak R, McGinty D. Sleep deprivation suppresses neurogenesis in the adult hippocampus of rats. Eur J Neurosci. 2005;22:2111–6. doi: 10.1111/j.1460-9568.2005.04376.x. [DOI] [PubMed] [Google Scholar]
  • 17.Taki Y, Hashizume H, Thyreau B, et al. Sleep duration during weekdays affects hippocampal gray matter volume in healthy children. Neuroimage. 2012;60:471–5. doi: 10.1016/j.neuroimage.2011.11.072. [DOI] [PubMed] [Google Scholar]
  • 18.Neylan TC, Mueller SG, Wang Z, et al. Insomnia severity is associated with a decreased volume of the CA3/dentate gyrus hippocampal subfield. Biol Psychiat. 2010;68:494–6. doi: 10.1016/j.biopsych.2010.04.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Watson C, Andermann F, Gloor P, et al. Anatomic basis of amygdaloid and hippocampal volume measurement by magnetic resonance imaging. Neurology. 1992;42:1743–50. doi: 10.1212/wnl.42.9.1743. [DOI] [PubMed] [Google Scholar]
  • 20.Fischl B. FreeSurfer. Neuroimage. 2012;62:774–81. doi: 10.1016/j.neuroimage.2012.01.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lao Z, Shen D, Xue Z, Karacali B, Resnick SM, Davatzikos C. Morphological classification of brains via high-dimensional shape transformations and machine learning methods. Neuroimage. 2004;21:46–57. doi: 10.1016/j.neuroimage.2003.09.027. [DOI] [PubMed] [Google Scholar]
  • 22.Ashburner J, Friston KJ. Voxel-based morphometry-the methods. Neuro-image. 2000;11:805–21. doi: 10.1006/nimg.2000.0582. [DOI] [PubMed] [Google Scholar]
  • 23.Klöppel S, Abdulkadir A, Jack CR, Jr, Koutsouleris N, Mourão-Miranda J, Vemuri P. Diagnostic neuroimaging across diseases. Neuroimage. 2012;61:457–63. doi: 10.1016/j.neuroimage.2011.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Publishing; 2000. text revision (DSM-IV-TR) [Google Scholar]
  • 25.Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia. 1971;9:97–113. doi: 10.1016/0028-3932(71)90067-4. [DOI] [PubMed] [Google Scholar]
  • 26.American Academy of Sleep Medicine. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Westchester, IL: American Academy of Sleep Medicine; 2007. [Google Scholar]
  • 27.Bastien CH, Vallières A, Morin CM. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2:297–307. doi: 10.1016/s1389-9457(00)00065-4. [DOI] [PubMed] [Google Scholar]
  • 28.Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiat Res. 1989;28:193–213. doi: 10.1016/0165-1781(89)90047-4. [DOI] [PubMed] [Google Scholar]
  • 29.Morin CM, Vallières A, Ivers H. Dysfunctional beliefs and attitudes about sleep (DBAS): validation of a brief version (DBAS-16) Sleep. 2007;30:1547–54. doi: 10.1093/sleep/30.11.1547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Broomfield NM, Espie CA. Towards a valid, reliable measure of sleep effort. J Sleep Res. 2005;14:401–7. doi: 10.1111/j.1365-2869.2005.00481.x. [DOI] [PubMed] [Google Scholar]
  • 31.Nicassio PM, Mendlowitz DR, Fussell JJ, Petras L. The phenomenology of the pre-sleep state: the development of the pre-sleep arousal scale. Behav Res Ther. 1985;23:263–71. doi: 10.1016/0005-7967(85)90004-x. [DOI] [PubMed] [Google Scholar]
  • 32.Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540–5. doi: 10.1093/sleep/14.6.540. [DOI] [PubMed] [Google Scholar]
  • 33.Beck AT, Steer RA. Beck Depression Inventory. San Antonio: The Psychological Corporation; 1987. [Google Scholar]
  • 34.Spielberger CD, Gorsuch RL, Lushene R, Vogg PR, Jacobs GA. Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists Press; 1983. [Google Scholar]
  • 35.Mugler JP, Brookeman JR. Three-dimensional magnetization-prepared rapid gradient-echo imaging (3D MP RAGE) Magnet Reson Med. 1990;15:152–7. doi: 10.1002/mrm.1910150117. [DOI] [PubMed] [Google Scholar]
  • 36.Fischl B, Dale AM. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc Natl Acad Sci USA. 2000;97:11050–5. doi: 10.1073/pnas.200033797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Fischl B, Salat DH, Busa E, et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 2002;33:341–55. doi: 10.1016/s0896-6273(02)00569-x. [DOI] [PubMed] [Google Scholar]
  • 38.Han X, Jovicich J, Salat D, et al. Reliability of MRI-derived measurements of human cerebral cortical thickness: the effects of field strength, scanner upgrade and manufacturer. Neuroimage. 2006;32:180–94. doi: 10.1016/j.neuroimage.2006.02.051. [DOI] [PubMed] [Google Scholar]
  • 39.Desikan RS, Segonne F, Fischl B, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 2006;31:968–80. doi: 10.1016/j.neuroimage.2006.01.021. [DOI] [PubMed] [Google Scholar]
  • 40.Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B. 1995;57:289–300. [Google Scholar]
  • 41.Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20:273–97. [Google Scholar]
  • 42.Brodersen KH, Ong CS, Stephan KE, Buhmann JM. The balanced accuracy and its posterior distribution. Proceedings of the 20th International Conference on Pattern Recognition, 2010; 2010. pp. 3121–4. [Google Scholar]
  • 43.Ashburner J. A fast diffeomorphic image registration algorithm. Neuro-image. 2007;38:95–113. doi: 10.1016/j.neuroimage.2007.07.007. [DOI] [PubMed] [Google Scholar]
  • 44.Spiegelhalder K, Regen W, Kyle SD, et al. Time will tell: a retrospective study investigating the relationship between insomnia and objectively defined punctuality. J Sleep Res. 2012;21:264–9. doi: 10.1111/j.1365-2869.2011.00970.x. [DOI] [PubMed] [Google Scholar]
  • 45.Spiegelhalder K, Kyle SD, Feige B, et al. The impact of sleep-related attentional bias on polysomnographically measured sleep in primary insomnia. Sleep. 2010;33:107–12. doi: 10.1093/sleep/33.1.107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Nissen C, Kloepfer C, Feige B, et al. Sleep-related memory consolidation in primary insomnia. J Sleep Res. 2011;20:129–36. doi: 10.1111/j.1365-2869.2010.00872.x. [DOI] [PubMed] [Google Scholar]
  • 47.Harvey AG. A cognitive model of insomnia. Behav Res Ther. 2002;40:869–93. doi: 10.1016/s0005-7967(01)00061-4. [DOI] [PubMed] [Google Scholar]
  • 48.Perlis ML, Merica H, Smith MT, Giles DE. Beta EEG activity and insomnia. Sleep Med Rev. 2001;5:363–74. doi: 10.1053/smrv.2001.0151. [DOI] [PubMed] [Google Scholar]
  • 49.Nofzinger EA, Buysse DJ, Germain A, Price JC, Miewald JM, Kupfer DJ. Functional neuroimaging evidence for hyperarousal in insomnia. Am J Psychiat. 2004;161:2126–8. doi: 10.1176/appi.ajp.161.11.2126. [DOI] [PubMed] [Google Scholar]
  • 50.Altena E, van der Werf YD, Sanz-Arigita EJ, et al. Prefrontal hypoactivation and recovery in insomnia. Sleep. 2008;31:1271–6. [PMC free article] [PubMed] [Google Scholar]
  • 51.Koolschijn PC, van Haren NE, Lensvelt-Mulders GJ, Hulshoff Pol HE, Kahn RS. Brain volume abnormalities in major depressive disorder: a meta-analysis of magnetic resonance imaging studies. Hum Brain Mapp. 2009;30:3719–35. doi: 10.1002/hbm.20801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Smith ME. Bilateral hippocampal volume reduction in adults with post-traumatic stress disorder: a meta-analysis of structural MRI studies. Hippocampus. 2005;15:798–807. doi: 10.1002/hipo.20102. [DOI] [PubMed] [Google Scholar]
  • 53.Steen RG, Mull C, McClure R, Hamer RM, Lieberman JA. Brain volume in first-episode schizophrenia: systematic review and meta-analysis of magnetic resonance imaging studies. Brit J Psychiat. 2006;188:510–8. doi: 10.1192/bjp.188.6.510. [DOI] [PubMed] [Google Scholar]
  • 54.Davatzikos C, Xu F, An Y, Fan Y, Resnick SM. Longitudinal progression of Alzheimer's like patterns of atrophy in normal older adults: the SPARE-AD index. Brain. 2009;132:2026–35. doi: 10.1093/brain/awp091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Klöppel S, Chu C, Tan GC, et al. Automatic detection of preclinical neurodegeneration: presymptomatic Huntington disease. Neurology. 2009;72:426–31. doi: 10.1212/01.wnl.0000341768.28646.b6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Koutsouleris N, Meisenzahl EM, Davatzikos C, et al. Use of neuro-anatomical pattern classification to identify subjects in at-risk mental states of psychosis and predict disease transition. Arch Gen Psychiatry. 2009;66:700–12. doi: 10.1001/archgenpsychiatry.2009.62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Vgontzas AN, Liao D, Pejovic S, et al. Insomnia with short sleep duration and mortality: the Penn State cohort. Sleep. 2010;33:1159–64. doi: 10.1093/sleep/33.9.1159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Vgontzas AN, Fernandez-Mendoza J, Bixler EO, et al. Persistent insomnia: the role of objective short sleep duration and mental health. Sleep. 2012;35:61–8. doi: 10.5665/sleep.1586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Feige B, Al-Shajlawi A, Nissen C, et al. Does REM sleep contribute to subjective wake time in primary insomnia? A comparison of polysomno-graphic and subjective sleep in 100 patients. J Sleep Res. 2008;17:180–90. doi: 10.1111/j.1365-2869.2008.00651.x. [DOI] [PubMed] [Google Scholar]
  • 60.Ohayon MM, Carskadon MA, Guilleminault C, Vitiello MV. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: developing normative sleep values across the human lifespan. Sleep. 2004;27:1255–73. doi: 10.1093/sleep/27.7.1255. [DOI] [PubMed] [Google Scholar]

RESOURCES