Abstract

Despite mounting reports about the negative effects of chronic occupational stress on cognitive functions, it is still uncertain whether and how this type of stress is associated with cerebral changes. This issue was addressed in the present MRI study, in which cortical thickness (Cth) and subcortical volumes were compared between 40 subjects reporting symptoms of chronic occupational stress (38 ± 6 years) and 40 matched controls (36 ± 6 years). The degree of perceived stress was measured with Maslach Burnout Inventory. In stressed subjects, there was a significant thinning of the mesial frontal cortex. When investigating the correlation between age and Cth, the thinning effect of age was more pronounced in the stressed group in the frontal cortex. Furthermore, their amygdala volumes were bilaterally increased (P = 0.020 and P = 0.003), whereas their caudate volumes were reduced (P = 0.040), and accompanied by impaired fine motor function. The perceived stress correlated positively with the amygdala volumes (r = 0.44, P = 0.04; r = 0.43, P = 04). Occupational stress was found to be associated with cortical thinning as well as with selective changes of subcortical volumes, with behavioral correlates. The findings support the hypothesis that stress-related excitotoxicity might be an underlying mechanism, and that the described condition is a stress related illness.

Introduction

Western societies are facing increasing reports of stress-related sickness among otherwise healthy and high-performing persons who report that they have not experienced any major negative life events or particular stress in early life (Ahola et al. 2006; Fernandez Torres et al. 2006; Rydmark et al. 2006; Copertaro et al. 2007, 2010). These persons describe having stereotyped symptoms, including memory and concentration problems, sleeplessness, diffuse aches, profound fatigue, irritability, anxiety, and a feeling of being emotionally drained, which they often attribute to occupational stress. Even though many individuals recover from the acute symptoms, the cognitive and emotional dysfunction as well as the increased sensitivity to stress often last for months, or years, forcing the affected individuals to work part-time, change jobs or retire early. The condition is often designated as occupational burnout, a label, which will be used operatively in the article. Its mechanisms are largely unknown, but there is reason to believe (Jovanovic et al. 2011; Backstrom et al. 2013; Blix et al. 2013) that this condition should be primarily investigated in the context of stress pathophysiology.

The predominant conceptual definition of stress, initially defined by Lazarus and Folkman (1987), suggests that stress may be thought of as resulting from an “imbalance between demands and resources.” This interpretation focuses on the transaction between people and their external environment and is known as the transactional model. The transactional model contains several core elements: The typical mode of behavior in subjects describing stereotyped symptoms from occupational stress seems to comprise an underestimation of the significance of occupational load, acceptance of high workload, and a feeling that the work situation can be controlled by increasing the working hours. This maladaptive coping effort and strategy results in a drastically reduced recuperation, sleep problems, fatigue, and subsequently to cognitive (Sandstrom et al. 2005; Jovanovic et al. 2011) and emotional dysfunction with impaired ability to modulate emotion (own unpublished data). While initially not appraising work as a stressor, but rather a positive challenge, they end up with a strong demand-control imbalance leading to emotional conflict with a presumable overactivation of the prefrontal networks (Egner et al. 2008) and perception of stress.

  • Primary appraisal, which is the evaluation of the significance of a stressor or threatening event.

  • Secondary appraisal—defined as evaluation of the controllability of the stressor and a person's coping resources.

  • Coping efforts—actual strategies used to mediate primary and secondary appraisals.

The major pathways of the physiological response to stress involve the autonomic nervous system as well as the hypothalamic–pituitary–adrenal (HPA) axis (Ulrich-Lai and Herman 2009). Although an alteration in the reactivity of the HPA system is believed to occur in subjects suffering from symptoms attributed to chronic occupational stress, this does not seem to be a consistent finding (Fries et al. 2009), as various studies have found normal (Mommersteeg, Heijnen et al. 2006; Langelaan et al. 2007), reduced (Pruessner et al. 1999; Moch et al. 2003; Mommersteeg, Keijsers, et al. 2006; Chida and Steptoe 2009), and elevated (Melamed et al. 1999; Grossi et al. 2003, 2005) cortisol levels after awakening. Because many stressed subjects report anxiety, sleeplessness, and poor initiative, their illness is frequently diagnosed as depression. It seems, however, to represent a separate construct, even if some symptoms (such as anxiety and attention and memory deficits) may overlap with depression. Only a minor portion of those suffering from symptoms related to occupational stress are helped by antidepressants (Asberg et al. 2010) and, in contrast to many patients with major depression (Friess et al. 2008), individuals with occupational stress have shown reduced cortisol and adreno-cortico hormone responses to the corticotropin-releasing hormone after dexametasone pretreatment (Rydmark et al. 2006; Wahlberg et al. 2009).

The described cognitive and emotional symptoms, as well as fatigue suggest that major limbic and paralimbic networks may be involved, which raises the question of whether structural changes occurred in these networks. In a recent study using positron emission tomography, we found that patients suffering from chronic occupational stress and burnout symptoms had significant reductions in the 5-HT1A receptor binding in 3 limbic structures: the hippocampus, the anterior cingulate cortex (ACC), and the anterior insular cortex (Jovanovic et al. 2011). In addition, there was a functional disconnection between the amygdala and the medial prefrontal cortex (mPFC), including the ACC, despite an absence of psychiatric comorbidity and major negative life events (Jovanovic et al. 2011). The locations of these changes, in several aspects, corresponded to the locations of structural changes detected through MRI in persons suffering from other stress-related conditions, such as stress in early life, repeated stressful negative life events, and post-traumatic stress disorder (PTSD) (Wang et al. 2000; Cohen et al. 2006; Pavlisa, Papa, Pavic, Pavlisa 2006; Bremner 2007; Bremner et al. 2008; Geuze et al. 2008; Kasai et al. 2008; Porcelli et al. 2008; van Harmelen et al. 2010; Papagni et al. 2011; Kuo et al. 2012; Morey et al. 2012; Pitman et al. 2012). This similarity is of interest for a couple reasons. First, since the pathophysiology of symptoms attributed to occupational stress is highly debated, there is a need for clarification on whether this condition is associated with any cerebral changes. Second, if structural changes similar to those described in other stress-related conditions are also linked to occupational stress, one may hypothesize that chronic psychosocial stress affects our brains in a rather stereotyped manner, regardless of the underlying cause, and that cerebral changes occur not only in response to exposure to extreme and life threatening situations, but could also be an effect of accumulated everyday stress. Indeed, in a recently published MR study (Blix et al. 2013), we found significant reductions in the relative volume of the caudate and putamen in subjects reporting cognitive impairment attributed to chronic occupational stress; using voxel-based morphometry, we also detected reductions in the gray-matter (GM) volume of the ACC and the mPFC. GM volume is, however, a composite and rather unspecific metric, and several recent studies show that the 2 lower order measures of GM volume, cortical thickness, and surface area are more appropriate to take into account when investigating factors regarding cortical morphology (Kruggel et al. 2003; Sowell et al. 2003). In the present MRI study, an entirely new group of stressed subjects and controls was investigated specifically with regard to cortical thickness (Cth) and surface area. The study also included an analysis of the structural volumes of the amygdala, the hippocampus, thalamus, and the cerebellum, in addition to the caudate, and putamen. Each of these structures is, according to previous data from animal experiments, involved in the processing of stress stimuli (Ryzhavskii et al. 2003; Ladefoged et al. 2004; Cohen et al. 2006; Filipovic et al. 2011; Yin et al. 2011; Leuner and Shors 2012). Changes in several of these structures have also been found in patients with early life trauma (Cohen et al. 2006; Baker et al. 2013) and in those with PTSD (Sapolsky 2000; Bonne et al. 2001; Pavlisa, Papa, Pavic 2006; Bremner et al. 2008; Filipovic et al. 2011; Papagni et al. 2011; Zhang et al. 2011). Together, these data suggest a direct link to stress that is independent of the specific type of stressor.

On the basis of previous experimental data, it was assumed that repetitive, stress-induced activation could lead to neuronal and dendritic damage in the structures involved. The underlying hypothesis was that subjects suffering from chronic occupational stress would show changes in cortical thickness, primarily in the mPFC and insular cortex. A further hypothesis was that structural volumes would be reduced in the amygdala, hippocampus, caudate, and putamen. Considering our previous observation of reduced basal ganglia volume among subjects suffering from occupational stress, we also specifically investigated fine motor function in this study, since this function reportedly requires that the processes of the basal ganglia are intact (Ciumas et al. 2010).

Materials and Methods

Subjects

Forty right-handed (Oldfield 1971) subjects (15 men, 25 women; age 38 ± 6 years, range 19–46 years; education 17 ± 3 years), who had been diagnosed as having had a “reaction to severe stress and an adjustment disorder” according to the International Classification of Diseases (ICD-10, F43) were recruited.

The selection was limited to subjects who attributed their illness to prolonged work-related stress, after working 60–70 h per week continuously over several years prior to the onset of symptoms. Inclusion criteria consisted of a characteristic symptom course of sleeplessness, diffuse aches, palpitations and fatigue, a subsequent onset of irritability, anxiety, memory and concentration problems, feeling of depersonalization, and reduced work capacity (confirmed by the employers) (Sandstrom et al. 2005; Rydmark et al. 2006). All of the subjects attributed their symptoms to chronic stress and had no other known etiology for their distress. All the subjects in the stressed group reported reduced sleep hours and fragmented sleep on the Karolinskas Sleep Questionnaire (Akerstedt et al. 2002).

Subjects were also required to have had a symptom duration of at least one year (their histories of stress-related burnout symptoms ranged from 1.5 to 3.5 years), to have been on sick leave (≥50%) for stress-related symptoms for a minimum of 6 months before entering the study, and to have an average stress-burnout score of ≥3.0 on the Maslach Stress-Burnout Inventory—General Survey (MBI-GS) (Schaufeli and Van Dierendonck 1995). This 7-point rating scale, ranging from 0 (never) to 6 (daily), consists of 3 subscales: exhaustion (5 items), cynicism (5 items), and lack of professional efficacy (6 items). When rating perceived stress, subjects were asked to take into consideration the last 6 months, and not only the actual time point. The average scores for Scandinavian populations are around 2 for MBI-GS (Ahola et al. 2006; Stenlund et al. 2007).

Subjects were excluded if they had a previous history of psychosis, personality disorder, major or bipolar depression, alcohol or substance abuse, chronic fatigue, chronic pain, fibromyalgia, or neurological or endocrine disease. Those who had experienced prominent stress factors in their private life or a major traumatic event at any time in their life, including sexual abuse, were also excluded. No daily medication was allowed during the 2 months prior to the study, except contraceptives. According to a review of their pharmacological treatment histories, none of them had taken drugs that are known to affect brain structure (e.g., psychopharmaca).

Forty healthy, right-handed, nonsmoking volunteers (15 females; age 36 ± 6 years, range 25–45 years; education 17 ± 3 years) with no history of chronic stress or heredity for neuropsychiatric disorders comprised the control group. The patient and control groups had similar gender distributions, and both were predominately female to accord with the female-dominated epidemiology of the condition studied (Ahola et al. 2006).

The 2 groups were matched for socioeconomic status assessed on the basis of years of education, type of occupation, and the level of management (employee, middle management responsibility, chief). All were white-collar workers, all were moderate alcohol consumers (on average 3 glasses of wine/week), and they rated similarly on the social support scale (Matsukura et al. 2002). The study was approved by the Ethics Committee at Karolinska Institutet, and written informed consent was received from each participant.

Before the interview, participants completed questionnaires in order to evaluate their stress symptoms and assess their previous life events (Deykin et al. 2001). In addition, the occurrence of major life events among the subjects was assessed through a clinical psychiatric interview based on the nonwork-related items of the Holmes and Rahe (1967) Scale. The participants were asked to answer yes or no to whether they had experienced any nonwork-related stressful life events (e.g., death of a relative or spouse, recent divorce, forced family relocation). They also completed the Karolinska Sleep Questionnaire. Subjects were excluded if they answered positively to having experienced such an event in their lives. Patients also received a medical screening (physical examination, test of thyroid and liver function). The possible presence of psychiatric disorders or personality disturbances were assessed according to the Diagnostic and Statistical Manual of the American Psychiatric Association, 4th Edition (DSM–IV), including the Structured Questionnaire for DSM-IV® Axis I and II (Structured Clinical Interview for DSM-IV®) a shortened version of (SCID-I, and II) (American Psychiatric Publishing, Inc., Arlington, TX, USA, 1997), along with a test for depression using the Montgomery–Asberg Depression scale (Montgomery et al. 1978).

Salivary Cortisol

Salivary cortisol was sampled in accordance with a previously established protocol (Karlamangla et al. 2013). Saliva sampling was chosen because the method is simple, noninvasive, and nonstressful; saliva samples have been shown to reflect the levels of the free fraction of cortisol in plasma (Galbois et al. 2010). Subjects were carefully instructed on how to collect the samples, which was done by placing Salivette cotton rolls (Sarstedt, Rommelsdorf, Germany) in their mouths for 2min. The samples were taken 7 times over the course of a weekday. The first sample (CS1) was collected immediately upon awakening in the morning, irrespective of time. The second sample (CS2) was collected 15 min later, before eating or brushing teeth, and the third sample (CS3) was collected 15 min after that. The fourth sample (CS4) was collected around noontime, before lunch. The fifth sample (CS5) was collected at about 3 PM, the sixth (CS6) at 8:00 PM, and the seventh (CS7) at bedtime, after having been at rest in bed for 15 min and before falling asleep. The samples were frozen (−18°C) until analyzed. The levels of salivary cortisol were measured by radioimmunoassay using the Spectria (125I)-coated tubes radioimmunoassay kit (Orion Diagnostica, FIN-02101 Espoo, Finland).

Magnetic Resonance Imaging

Data Acquisition

Magnetic resonance imaging data was acquired on a 3-Tesla MRI medical scanner (Discovery 3T GE-MR750, General Electric, Milwaukee, WI, USA) equipped with a 32-channel and/or 8-channel phased array receiving coil. 3D T1-weighted SPGR images were acquired with 1-mm3 isotropic voxel size (TE = 3.1 ms, TR = 7.9 ms, TI = 450 ms, FoV = 24 cm, 176 axial slices, flip angle of 12°). Other MR sequences (not used in the present study) included resting-state functional MRI, performed with a gradient-echo pulse sequence using a voxel size of 3 × 3 mm (TE = 30 ms, TR = 2500 ms, FoV = 28.8 cm, 44 interleaved axial slices, 3 mm thickness, flip angle of 90°). In addition, multislice DTI (not used in the present analyses) was performed using an echo-planar imaging sequence with 2 × 2 mm in plane resolution, [FoV = 23 cm, 60 interleaved axial slices, thickness = 2.9 mm, TE = 83.7 ms, TR 8000 ms, 60 diffusion gradient directions (b = 1000), flip angle of 90°]. Finally, there was a clinical sagittal FLAIR: TE/TR = 117/8000, TI = 2255, ETL = 140, ARC acceler. R = 2 × 2 (slice, phase), FoV: 27 cm, 224 × 224, slice thickness, 1.2 mm.

Cortical Thickness and Surface Area

The MR volumes were processed using FreeSurfer software version 5.1 (www.surfer.nmr.mgh.harvard.edu). To calculate the surface-based anatomical measures, models of white-matter (WM) and GM surfaces were reconstructed from MR volumes. The thickness was then measured as the distance between the WM and the GM surfaces at each vertex. The vertices were arranged in ∼1 mm spacing, which allows measuring cortical thickness at up to 160 000 vertices in each hemisphere with submillimeter precision. Before the surfaces were reconstructed, several preprocessing steps were done (Fischl and Dale 2000). First, the volume was registered to the Talairach atlas using affine transformation. Second, to bring about intensity normalization, the bias field was computed and used to correct intensity variability across the image caused by radiofrequency field inhomogenieties and susceptibility artifacts. Third, the skull (nonbrain) tissue was removed by deforming a tessellated ellipsoidal template into the shape of the inner surface of the skull. Fourth, the WM hemispheres were constructed; during this procedure, the overlap in intensities between WM and GM was taken into account as well as the fact that the WM/GM borders should be planar due to the laminar structure of the cortex. Fifth, the cutting planes were chosen in such a way as to separate the hemispheres from each other as well as to remove the cerebellum and brain stem. Finally, the WM surface was generated by covering the filled WM hemisphere with triangles (tessellation) and smoothing it to follow the intensity gradients between WM and GM; then, the GM (pial) surface was generated by expanding the WM surface to follow the intensity gradients between the GM and CSF.

The reconstruction of the MRI images was inspected visually at several stages: after the Talairach transformation, after the skull striping, and, finally, after the surfaces had been built and the volumes labeled. At each of these stages, the necessary corrections were made, including correcting erroneous skull striping by adjusting watershed parameters or by manually editing out the skull tissue, and adding control points to normalize intensity for erroneous WM surface reconstruction.

Tests of Group Differences in Cortical Thickness and Surface Area

The study was focused on investigating cortical thickness, but as the FreeSurfer program also calculates surface area, we were able to compare this factor between the 2 populations. All of the possible differences between the groups were evaluated for each vertex, using age as the nuisance variable, and after employing Monte Carlo correction (5000 permutations). This correction method for multiple comparisons generates random noise fields and detects the clusters that appear at specific size and probability thresholds. In our study, after 5000 of such iterations, a frequency of how often a simulated cluster's value exceeded the value from the true data analysis was computed, and this frequency was used to indicate the results which were significant at the level corrected for multiple comparisons (here set to P < 0.05). Although the groups were age and sex matched, it was, on the basis of previous publications (Sowell et al. 2007), assumed that age could be a factor influencing Cth and surface area. We also specifically tested if the 2 groups differed in regard to the effect of age on Cth, by employing age as covariate of interest. A different-offset-different-slope model was used (the slope of the line that represents thickness and surface area as a function of age was set to be different for the groups being compared, and the offset of these 2 lines was allowed to vary during fitting to the data). Within-group analyses were carried out using the same-offset-same-slope model.

Segmentation of the Subcortical Volumes

Subcortical segmentation generated with FreeSurfer was used to calculate the volumes of 6 subcortical brain structures: the amygdala, hippocampus, caudate, putamen, thalamus, and cerebellum. When needed, the segmented brain structural masks were modified manually. They were first registered to original gray-scale images, and then refined with 1 pixel erosion to eliminate possible partial volume artifacts and alignment errors. The mean signal intensity (SI) was finally calculated for each of the studied subcortical brain structures by averaging the SI of the voxels extracted from the original gray-scale images based on the segmentation defined volumes of interest (VOIs) using FreeSurfer software (www.surfer.nmr.mgh.harvard.edu), according to the standard procedure (Fischl et al. 2002, 2004). Finally, the structural volumes were labeled based on 3 types of probabilities at each voxel: prior (before observing a volume) probability that a voxel belongs to each of the label classes, probability that a given voxel belongs to the label given the classification of its neighboring points, and probability distribution function of intensity (for volume-based labeling). The same persons analyzed all the subcortical volumes. Ratios between the respective VOI and the total intracranial volume (TIV) were finally entered into the statistical analyses. After ensuring that the data were normally distributed, group comparisons of relative structural volumes (VOI/TIV) were carried out with a one-way ANOVAs using the individual relative values of the 2 homologous VOIs for each type of structure as input values (P < 0.05, with the Bonferroni correction for the thalamus and cerebellum, but not for the amygdala, hippocampus, and the basal ganglia, for which we had specific hypotheses). The aforementioned analyses were carried out with PASW Statistics 21 (SPSS, Inc., Chicago, IL, USA).

Possible Relationship Between Cth and the MBI-GS and MADRAS Scores

A possible relationship between stress and regional Cth was exploratively assessed using stress score (MBI-GS) as the covariate of interest and age as the nuisance variable (FreeSurfer software, P < 0.05, Monte Carlo correction). The same approach was used to investigate whether there was a relationship between Cth and MADRAS scores. The possible effects of stress on the subcortical volumes that differed between the 2 groups of subjects were investigated using Pearson's linear regression (P < 0.05) for the MBI-GS scores and the respective VOI/TIV.

Test of the Psychomotor Function

Psychomotor function was investigated using the grooved pegboard test, as described in one of our previous studies (Ciumas et al. 2010). The Grooved Pegboard test is a fine motor task that requires manipulative dexterity, speed, and complex visual–motor coordination in order to place 25 pegs, one at a time, into 25 keyhole-shaped holes with varying orientations that are randomly positioned on a 5-by-5 matrix. The pegs had to be rotated into position to match the hole before they were inserted, which was to be done as quickly as possible and in a prescribed order. Subjects were required to place the pegs from left to right on the board when using their right hand, and to go in the opposite direction when using their left hand. The criterion variable was the time needed to complete the task.

Results

General Demographics

The major demographic data are presented in Table 1. The groups were matched for age, sex, and years of education, but differed significantly with respect to stress scores (P < 0.001) and MADRAS scores (P < 0.001). However, none of the stressed subjects were deemed to be depressed according to the MINI questionnaire, and the mean MADRAS score in the stressed group was still within the nondepression range (<20). There was no group difference with respect to cortisol levels, either overall (P = 0.780; F = 0.078) or at any specific time point (Table 1).

Table 1

Demographics

 Stressed subjects (n = 40) Controls (n = 40) P, F values 
Age (years) 37.9 ± 6.4 36.1 ± 5.9 P = 0.17, F = 1.9 
Education (years) 16.6 ± 2.8 16.8 ± 3.1 P = 0.21, F = 1.6 
MBI-GS (score) 3.9 ± 0.6 2.6 ± 0.4 P = 0.00, F = 85.9 
 Exhaustion 4.4 ± 1.1 1.3 ± 1.0 P = 0.00, F = 300.4 
 Cynicism 3.2 ± 1.3 1.3 ± 0.9 P = 0.00, F = 85.9 
MADRAS (score) 16.8 ± 5.1 3.3 ± 2.9 P = 0.00, F = 213.4 
Cortisol sample 1 13.7 ± 11.0 14.2 ± 6.2 P = 0.3, F = 0.8 
Cortisol sample 2 21.3 ± 13.5 19.3 ± 9.2 P = 0.4, F = 0.7 
Cortisol sample 3 21.3 ± 11.3 22.3 ± 11.6 P = 0.7, F = 0.1 
Cortisol sample 4 8.7 ± 10.8 7.8 ± 4.6 P = 0.6, F = 0.3 
Cortisol sample 5 4.9 ± 2.4 6.6 ± 9.4 P = 0.3, F = 1.0 
Cortsiol sample 6 3.6 ± 4.6 3.7 ± 4.7 P = 0.8, F = 0.5 
Cortisol sample 7 2.2 ± 1.9 2.5 ± 3.9 P = 0.7, F = 1.2 
 Stressed subjects (n = 40) Controls (n = 40) P, F values 
Age (years) 37.9 ± 6.4 36.1 ± 5.9 P = 0.17, F = 1.9 
Education (years) 16.6 ± 2.8 16.8 ± 3.1 P = 0.21, F = 1.6 
MBI-GS (score) 3.9 ± 0.6 2.6 ± 0.4 P = 0.00, F = 85.9 
 Exhaustion 4.4 ± 1.1 1.3 ± 1.0 P = 0.00, F = 300.4 
 Cynicism 3.2 ± 1.3 1.3 ± 0.9 P = 0.00, F = 85.9 
MADRAS (score) 16.8 ± 5.1 3.3 ± 2.9 P = 0.00, F = 213.4 
Cortisol sample 1 13.7 ± 11.0 14.2 ± 6.2 P = 0.3, F = 0.8 
Cortisol sample 2 21.3 ± 13.5 19.3 ± 9.2 P = 0.4, F = 0.7 
Cortisol sample 3 21.3 ± 11.3 22.3 ± 11.6 P = 0.7, F = 0.1 
Cortisol sample 4 8.7 ± 10.8 7.8 ± 4.6 P = 0.6, F = 0.3 
Cortisol sample 5 4.9 ± 2.4 6.6 ± 9.4 P = 0.3, F = 1.0 
Cortsiol sample 6 3.6 ± 4.6 3.7 ± 4.7 P = 0.8, F = 0.5 
Cortisol sample 7 2.2 ± 1.9 2.5 ± 3.9 P = 0.7, F = 1.2 

Age and education are expressed in years; MBI-GS is a questionnaire to score perceived work-related stress. Raw 3 indicates the mean total score, raw 4–5 the subscores for the exhaustion and cynicism. MADRAS, Montgomery–Asberg Depression Scale. There was no overall group difference in cortisol levels (P = 0.780; F = 0.078, repeated-measure ANOVA). Time of the day for cortisol samples: Sample 1: 06.30–07.30; Sample 2: 15 min after; Sample 3: 30 min after sample 1; Sample 4: 12.00–13.00; Sample 5: 15.00–16.00; Sample 6: 20.00–21.00. Sample 7: 22.30–23.30.

Cortical Thickness

In stressed subjects, the mPFC was thinner than in controls (vertex maximum, −log10(P) = −3.6, size 10.0 cm2, Talairach's coordinate −18 36 −16 (Fig. 1). The average difference in Cth in this region was 0.12 cm2, the mean F-value according to the F-map was 7.6 (the mean Cth was 2.75 ± 0.22 cm2 in stressed subjects and 2.87 ± 0.21 cm2 in controls). This difference remained when adding the MADRAS scores as the covariate of no interest. Furthermore, when using MADRAS scores as covariate of interest and merging both study groups, no correlation with Cth was detected in this region. There was an inverse correlation between MADRAS scores and Cth in several other regions, but only one cluster (vertex maximum –log10(P) = −4.0, size 7.8 cm2, Talairach's coordinate −60 −15 3) survived the multiple comparison correction.

Figure 1.

Clusters showing significant reductions in Cth in stressed subjects. The contrast (stressed subjects – controls) was calculated at P < 0.05 corrected for multiple comparisons (Monte Carlo permutation), using age and gender as the covariates of no interest. The projection of cerebral hemispheres (MR images of the Freesurfer atlas) is standardized. Scale is logarithmic and shows –log10(P). Warm colors indicate positive contrasts (higher values in stressed subjects), cool colors negative contrasts (lower values in stressed subjects).

Figure 1.

Clusters showing significant reductions in Cth in stressed subjects. The contrast (stressed subjects – controls) was calculated at P < 0.05 corrected for multiple comparisons (Monte Carlo permutation), using age and gender as the covariates of no interest. The projection of cerebral hemispheres (MR images of the Freesurfer atlas) is standardized. Scale is logarithmic and shows –log10(P). Warm colors indicate positive contrasts (higher values in stressed subjects), cool colors negative contrasts (lower values in stressed subjects).

Contrary to our hypothesis, no direct correlation between Cth and MGI-GS scores was detected when using the explorative analysis and MGS-GI scores as covariate of interest in relation to Cth. When restricting the search space to the mPFC region where significant thinning was found in the stressed group, and applying Person's correlation analysis, r was −0.2, and P = 0.079. No significant group difference was detected with respect to the surface area.

Post hoc Analyses of Cortical Thickness

Although there was no significant age difference between the groups, the age range was relatively high. It was, therefore, of interest to investigate how Cth covaried with age across the entire study population, and also, if there were any regions in which this covariation differed between the 2 study groups. In a separate FreeSurfer analysis, age was, therefore, employed as covariate of interest using the entire study group, and with the entire brain as search space. There were several clusters showing a significant inverse correlation with age (primarily in the parietal and frontal lobes, and to a minor extent in the occipital lobe, Supplementary Fig. 1 and Table 1). Furthermore, group comparison in regard to this covariation showed a significant difference in a left prefrontal cluster (vertex maximum –log10(P) = −3.6, size 11.2 cm2, Talairach's coordinate −17 37 −15). Hence, whereas age was negatively correlated with Cth in several regions this inverse correlation was significantly more pronounced among stressed subjects, possibly suggesting an age-by-stress interaction. There were no regions for which the correlation between age and Cth was more pronounced in controls.

Several more recent studies have shown that functional networks of the brain have an intrinsically cohesive modular structure in that the connections between regions are much denser within each module than between them. The modules are mainly composed of functionally, as well as anatomically, related brain regions and can be identified by maps of covariance. This is of importance for investigations of clinical populations since maps of covariance, such as for Cth, may vary as a function of disease processes, and, thus, differ between patients and controls (He et al. 2008). Therefore, it was of particular interest to investigate whether the mPFC region where the cortex was thinner in the burnout group, also had a different pattern of cortico-cortical Cth covariations in this group. Using the left mPFC region with significant group difference in Cth as a seed region of interest, it was found that the covariation was significantly stronger in the stressed group in the primary cortical projections to the mPFC (Petrides and Pandya 1999; the orbitofrontal and rostro-medial frontal cortex, the ACC, and the insular cortex, and on the right side also the supramarginal and fusiform gyrus; Table 2, Figure 2). In addition, the covariation in parts of the seed region itself was significantly more pronounced in the stressed group. No regions with greater covariation among controls were detected.

Table 2

Significant group difference in cortico-cortical covariations from the mPFC seed region

Regions Left mPFC seed
 
Maximum vertex-wise −log10(PCluster size, cm2 Talairach coordinates 
L Orbitofrontal cortexa 10.7 55.0 −19 17 −17 
L Caudal middle frontal cortex 3.6 10.2 −38 3 44 
L Superior frontal cortex 3.7 22.9 −7 28 49 
R Superior frontal cortex 6.9 40.1 9 28 33 
L Insular cortex 3.8 9.1 −29 21 4 
R Insular cortex 2.8 12.7 38 −10 4 
R Supramarginal cortex 4.1 8.2 58 −25 27 
R Fusiform gyrus 3.6 12.5 33 −59 −12 
Regions Left mPFC seed
 
Maximum vertex-wise −log10(PCluster size, cm2 Talairach coordinates 
L Orbitofrontal cortexa 10.7 55.0 −19 17 −17 
L Caudal middle frontal cortex 3.6 10.2 −38 3 44 
L Superior frontal cortex 3.7 22.9 −7 28 49 
R Superior frontal cortex 6.9 40.1 9 28 33 
L Insular cortex 3.8 9.1 −29 21 4 
R Insular cortex 2.8 12.7 38 −10 4 
R Supramarginal cortex 4.1 8.2 58 −25 27 
R Fusiform gyrus 3.6 12.5 33 −59 −12 

Statistical threshold is P < 0.05, corrected for multiple comparisons (according to Monte Carlo permutations). The individual mean Cth in the left mPFC seed region (where the cortex was significantly thinner in stressed subjects) was used as covariate of interest. The Talairach's coordinates indicate location of maximum vertex for the cluster. “Region” column describes the coverage of the respective cluster. Positive Maximum vertex-wise −log10(P) values indicate significantly greater covariation in stressed subjects compared with controls. aCovers a portion of the anterior cingulate cortex. R, right; L, left.

Figure 2.

Group difference (stressed subjects – controls) in cortico-cortical covariation from the left mPFC (the region showing a significantly thinner cortex in the stressed group), corrected for multiple comparisons (Monte Carlo permutation), and using age as the covariate of no interest. The projection of cerebral hemispheres (MR images of the Freesurfer atlas) is standardized. Scale is logarithmic and shows –log10(P). Warm colors indicate positive contrasts (higher values in stressed subjects), cool colors negative contrasts.

Figure 2.

Group difference (stressed subjects – controls) in cortico-cortical covariation from the left mPFC (the region showing a significantly thinner cortex in the stressed group), corrected for multiple comparisons (Monte Carlo permutation), and using age as the covariate of no interest. The projection of cerebral hemispheres (MR images of the Freesurfer atlas) is standardized. Scale is logarithmic and shows –log10(P). Warm colors indicate positive contrasts (higher values in stressed subjects), cool colors negative contrasts.

Subcortical Volumes

Contrary to the hypothesis, the stressed subjects showed a significantly larger amygdala volume, bilaterally but particularly on the right side, compared with controls (Tables 3 and 4). Furthermore, there was a significant and positive linear correlation between the individual MBI-GS scores and the relative volume of the amygdala on both sides (Pearson's coefficient = 0.435, P = 0.038 for the left amygdala; Pearson's coefficient = 0.427, P = 0.044 for the right amygdala)—thus, the larger the amygdala volume, the higher the stress score. This covariation remained when restricting the correlation analysis to the stressed group; a significant covariation was not found when restricting the analysis to the control group (calculation using mean of the right and left relative amygdala volume; r = 0.604, P = 0.000, and r = −0.188, P = 0.122, respectively), see Supplementary Figure 2. Another structure for which a group difference was found is the caudate, as the stressed group showed a significant volume reduction in this area (Table 3). Notably, the relative caudate volumes were inversely correlated with the MBI-GS exhaustion scores for both sides (Pearson's coefficient = −0.288; P = 0.010 for the left caudate; Pearson's coefficient = −0.237, P = 0.036 for the right caudate; results from the corresponding covariation subanalysis investigating each group separately, failed to reach the significance level). The significance threshold for the aforementioned correlation analyses was 0.05, and no correction for multiple comparisons was applied, as a covariation with the degree of perceived stress was hypothesized in the regions showing altered structural volumes among the stressed subjects. Group differences in the amygdala volumes remained when using the MADRAS scores as the covariate of no interest (P = 0.003, F = 6.33, mean of both sides), whereas the P value for caudate comparisons was subsignificant (P = 0.120, F = 2.09, mean of both sides). No significant group difference was detected in the relative structural volume of the hippocampus, putamen, thalamus, or cerebellum. The values of structural volumes are presented in Table 3.

Table 3

Structural volumes

Structural volumes (cm3Stressed subjects Controls P, F values 
L Caudate volume 3.9 ± 0.4 4.1 ± 0.4 P = 0.04, F = 4.5* 
R Caudate volume 3.9 ± 0.4 4.2 ± 0.4 P = 0.04, F = 4.3* 
L Putamen volume 4.7 ± 0.5 4.6 ± 0.6 P = 0.51, F = 0.44 
R Putamen volume 4.7 ± 0.5 4.6 ± 0.6 P = 0.29, F = 1.12 
L Hippocampus volume 4.0 ± 0.3 4.1 ± 0.4 P = 0.741, F = 0.1 
R Hippocampus volume 4.1 ± 0.4 4.1 ± 0.4 P = 0.31, F = 1.0 
L Amygdala 1.8 ± 0.2 1.7 ± 0.2 P = 0.018, F = 5.8* 
R Amygdala 2.0 ± 0.2 1.8 ± 0.2 P = 0.003, F = 9.5** 
L Thalamus volume 7.0 ± 0.7 7.2 ± 0.7 P = 0.58, F = 0.30 
R Thalamus volume 7.0 ± 0.7 7.2 ± 0.8 P = 0.63, F = 0.23 
Cerebellum 136.1 ± 6.7 136.2 ± 6.6 P = 0.42, F = 0.66 
TIV volume 1435.1 ± 187.5 1473.4 ± 158.0 P = 0.325, F = 0.98 
Structural volumes (cm3Stressed subjects Controls P, F values 
L Caudate volume 3.9 ± 0.4 4.1 ± 0.4 P = 0.04, F = 4.5* 
R Caudate volume 3.9 ± 0.4 4.2 ± 0.4 P = 0.04, F = 4.3* 
L Putamen volume 4.7 ± 0.5 4.6 ± 0.6 P = 0.51, F = 0.44 
R Putamen volume 4.7 ± 0.5 4.6 ± 0.6 P = 0.29, F = 1.12 
L Hippocampus volume 4.0 ± 0.3 4.1 ± 0.4 P = 0.741, F = 0.1 
R Hippocampus volume 4.1 ± 0.4 4.1 ± 0.4 P = 0.31, F = 1.0 
L Amygdala 1.8 ± 0.2 1.7 ± 0.2 P = 0.018, F = 5.8* 
R Amygdala 2.0 ± 0.2 1.8 ± 0.2 P = 0.003, F = 9.5** 
L Thalamus volume 7.0 ± 0.7 7.2 ± 0.7 P = 0.58, F = 0.30 
R Thalamus volume 7.0 ± 0.7 7.2 ± 0.8 P = 0.63, F = 0.23 
Cerebellum 136.1 ± 6.7 136.2 ± 6.6 P = 0.42, F = 0.66 
TIV volume 1435.1 ± 187.5 1473.4 ± 158.0 P = 0.325, F = 0.98 

P-values for structural volumes were based on calculations of ratios between the respective structural volume and the TIV.

*P < 0.05; **P < 0.01.

TIV, total intracranial volume.

Table 4

Results of the pegboard test

 Stressed subjects (n = 40) Controls (n = 40) P, F values 
Peg right (s) 62.8 ± 9.4 57.6 ± 6.5 P = 0.006, F = 8.045** 
Peg left (s) 70.9 ± 11.7 66.2 ± 11.3 P = 0.074, F = 3.328 
 Stressed subjects (n = 40) Controls (n = 40) P, F values 
Peg right (s) 62.8 ± 9.4 57.6 ± 6.5 P = 0.006, F = 8.045** 
Peg left (s) 70.9 ± 11.7 66.2 ± 11.3 P = 0.074, F = 3.328 

Peg scores: time needed to put all the pegs in the pegboard. All the subjects were right handed.

*P < 0.05; **P < 0.01.

Stressed subjects also showed impaired performance for the pegboard task when using the dominant (right) hand, and required significantly longer time to position the pegs (P = 0.006, F = 8.1). There was a similar trend for the left hand performance, but without passing the significance level (P = 0.074, F = 3.3) (Table 4). The performance in the pegboard task (time) did not correlate with the relative volume of the caudate or putamen.

Discussion

Cognitive impairment is frequently reported among subjects with chronic occupational stress. However, to the best of our knowledge, this is the first study to investigate whether occupational stress may be linked to changes in cortical thickness and structural volumes in regions considered to process psychosocial stress stimuli. It was found that the mPFC was significantly thinner in stressed subjects. Furthermore, the effect of age on cortical thinning in the mPFC was significantly more pronounced among stressed subjects. An additional finding was the increase in amygdala volume and the decrease in caudate volume in the stressed group. Both of these volume changes were related to the degree of perceived stress and caudate changes were accompanied by impaired performance in tests of fine motor function.

All of the aforementioned regions constitute parts of the limbic and paralimbic networks, and the detected changes are in accordance with a large body of animal data suggesting that stress affects this distributed network. The present data are unique in that they show regionally differentiated structural changes within this network in relation to everyday occupational stress. The present study advances the little that is known about the cerebral pathophysiology of this condition in several ways. First of all, it provides measurements of amygdala volume and shows that there is an association between an enlargement of this structure and higher scores of perceived stress. The amygdala is the first relay station in the processing of psychosocial stress stimuli (Koenigs and Grafman 2009), and animal experiments show that severe and/or prolonged stress is associated with a glutamate-induced elevation of brain-derived neurotrophic factor (BDNF), an increased arborization, and dendritic hypertrophy in the basolateral amygdala (Arendt et al. 2012; Boyle 2013). Similar mechanisms may also operate in humans.

Previous structural neuroimaging studies of stress have mainly been restricted to PTSD and early traumas, and seem inconclusive with regard to amygdala volume (Woon and Hedges 2009), with some finding no effect of diagnosis (Gurvits et al. 1996; Lindauer et al. 2004; Chao et al. 2013) and others finding either smaller amygdala volume or enlarged amygdala volume (Lupien et al. 2011; Weber et al. 2013) in patients. One reason for the disparate results is that these studies were often based on small samples and different types of trauma in different studies, which could be different in nature even among the same study group. In contrast, the present data were collected from a population that was homogenous with respect to stress type and exposure, and that had no comorbid conditions. These characteristics increase the reliability of the finding in this study that amygdala enlargement occurred among stressed persons.

A second finding, which suggests that occupational burnout is a limbic condition is the thinning of the mPFC. Considering that Cth and cortical area are subcomponents of GM volume, this finding advances our recent observation of reduced GM volume in this region (Blix et al. 2013), by helping narrow the number of possible morphological correlates (see next paragraph). The mPFC is a key site for the modulation of stress stimuli (Arnsten 2009), and numerous animal studies (Roozendaal et al. 2004), as well as studies of PTSD patients (Gilboa et al. 2004; Geuze et al. 2008; Koenigs and Grafman 2009; Lyoo et al. 2011) suggest that this structure is affected in individuals with chronic stress. The present observation that cortico-cortical covariation from the mPFC seed region to the insular cortex, the orbitofrontal and rostro-medial frontal cortex was greater in stressed subjects further underlines the limbic involvement of the occupational burnout condition. This observation is of particular interest because the aforementioned regions are axonally connected (Petrides and Pandaya 1988), and the stronger Cth correlation (Lerch et al. 2008) may confer to a mutually trophic effect and modulation/shaping of these connected regions (Pezawas et al. 2004) by an external factor, such as stress.

The detected caudate reduction and its link to perceived stress confirms our recent finding in a different sample of subjects suffering from occupational stress, and adds to it by demonstrating a functional correlate in terms of impaired psychomotor function.

Possible Mechanisms Underlying Changes in Cth and Subcortical Volumes

The exact underpinnings of the described limbic changes can at present only be speculated about. Because the study was cross-sectional, it is difficult to know whether the detected reductions represent neurotoxic effects of stress, effects of other factors, such as nitric oxide, or a pre-existing condition that could have rendered the brain more vulnerable to the development of pathological stress responses. Owing to the strict selection criteria, it is, nevertheless, possible to exclude potential confounding factors such as major life traumas, psychiatric premorbidity, chronic pain, and pharmacological treatment. Considering the congruence with data from animal experiments and some (Lyoo et al. 2011; Papagni et al. 2011), albeit not all (Bonne et al. 2001; De Bellis et al. 2001), reports from longitudinal studies of patients experiencing stressful life events, or PTSD, we find it more likely that the present findings reflect effects of chronic occupational stress. The observed correlation patterns between the stress scores and the relative amygdalae and caudate volumes support this notion, although the latter correlation was weak.

The study offered no option to directly examine the cellular correlates of the neuroimaging differences identified. Cortical thickness reflects cell-packing density, cell size, the number of cortical neurons, and the dendritic length (Kruggel et al. 2003), factors, which also contribute to subcortical volumes. The molecular underpinnings of the observed morphological changes will primarily be discussed in relation to neuronal modulation by stress. Two mediators of this modulation deserve particular attention—the glucocorticoids and the glutamate. Their effects have been studied in animal models in detail (Conrad 2008; Leuner and Shors 2012) and involve dendritic retraction, neurotoxicity, and, in some cases, even apoptosis (Bremner et al. 2008). Both glutamate and glucocorticoids have neuronal influences on their own, but are also reported to interact (Magarinos and McEwen 1995; Brown et al. 2010). Circulating glucocorticoids interact with various neurotransmitters (McEwen 2000a), and chronic stress in tree shrews is found to reduce the number of dopamine transporter binding sites (Bmax) in the caudate nucleus and the putamen (Isovich et al. 2000). An excess of glucocorticoids is reported to decrease the proliferation of hippocampal and prefrontal neurons, whereas adrenal ectomy is reported to increase it (Wellman 2001; Wong and Herbert 2004). Experiments with betamethasone infusion in fetal sheep show a glucocorticoid-related loss of synaptic density in the frontal neocortex, caudate, putamen, and hippocampus (Conrad et al. 1999; Colberg et al. 2004). However, while a possible impairment in cortisol homeostasis may have had an impact on the presently observed cerebral changes at some intermediate step of the stress-induced chain of events, the present data do not provide direct support for such a mechanism. As in several previous studies of occupational stress, neither the pattern of diurnal cortisol nor the levels at separate time points showed any difference compared with controls (Table 1).

This brings attention to the second stress-associated factor, which could have contributed to the observed changes—glutamate. Data from animal experiments show that stress causes an enhanced release of glutamate, and that a stress-related elevation of extracellular glutamate levels induces retraction in the spines in stress-targeted regions, such as the mPFC, the anterior cingulate, and the basal ganglia (Lowy et al. 1993; Magarinos and McEwen 1995; McEwen et al. 1997; Iijima et al. 2007; Hunter et al. 2009). Interestingly, in the amygdala, stress seems to lead to an increase in neurotrophic factors such as BDNF, and to enhance dendritic outgrowth. In the functional and anatomical projection areas of the amygdala, the stress-related glutamatergic excess is, on the other hand, reported to lead to reduced BDNF and shrinkage of dendrites (Arendt et al. 2012; Boyle 2013). These degenerative events are particularly pronounced in the mPFC (Roozendaal et al. 2004; Brown et al. 2005; Radley et al. 2008; Arnsten 2009; Leuner and Shors 2012). This is of interest because the mPFC exerts a strong negative control over stress pathways. GABAergic signals from the mPFC to the amygdala lead to a repression of the HPA axis. This provides a basis for one possible scenario for the present condition, in which stress-mediated neurotoxic damage to the mPFC, due to high glutamate, cortisol, or the combination of both (see Magarinos and McEwen 1995), has led to impaired prefrontal inhibition of the amygdala (Roozendaal et al. 2004). This would then have provided a context for a vicious circle with a further enhancement of amygdala excitation and subsequent changes along the networks connected to the amygdala (the mPFC, the basal ganglia, the hippocampus, and the insular, anterior cingulate, and orbitofrontal cortices). The presently reported findings among subjects suffering from occupational stress fit into this model, with the increase in amygdala volume and reduction in caudate volume, the thinning of the mPFC, and the more pronounced cortico-cortical covariation between the mPFC and insular cortex (Wang et al. 2007; Liston et al. 2009; Goldstein et al. 2010). It is also compatible with the notion that both the putamen and caudate receive powerful glutamatergic input from the amygdala (McEwen 2000b) and are susceptible to excitotoxicity (Chen et al. 1995; Bernal et al. 2000). This strengthens reiterating the hypothesis presented in our initial stress-related publication (Jovanovic et al. 2011) that repeated stress stimuli could cause excitotoxic, apoptotic, and/or intermediate forms of neuronal death (Bengzon et al. 1997) with atrophy in humans.

One observation worth commenting on is that the hippocampus was seemingly unaffected. Although at odds with animal data showing stress-related dendritic shrinkage and cell death in the hippocampus (McEwen 1994), the finding is congruent with previous volumetric studies among subjects with occupational stress (Sandstrom et al. 2005; Blix et al. 2013), and some (Bonne et al. 2001; Villarreal et al. 2002; Filipovic et al. 2011; Papagni et al. 2011; Chao et al. 2013), but not all (De Bellis et al. 2001; Landre et al. 2010), studies based on volumetric measurements of patients with PTSD and life traumas. One possible, yet speculative, explanation for this discrepancy is that the effects of stress on the hippocampus could be gender-related. Experimental data suggest that dendritic spine density in the hippocampus may respond in opposite directions to the same stressful event in male and female rats (Shors et al. 2001). A post hoc, within-same-sex group comparison of the present material failed to show any difference in the hippocampal volume. However, given the relatively limited size of the study group, this does not rule out the possibility that gender may have been a factor affecting the hippocampal changes that occurred in response to stress—a notion which deserves its own, separate study.

Occupational Stress and Depression

One important question worth discussing is whether and to what extent the present findings could reflect depression. For several reasons, we find this possibility to be unlikely. First of all, none of the stressed subjects were judged to be depressed according to the psychiatrist in charge and the SCID ratings. Although the depression scores were significantly higher among the stressed group, for those subjects who had high MADRAS scores, the only items, which contributed to these scores were anxiety and poor sleep, which does not necessarily implicate depression. Second, the detected group differences did not show an interaction with the MADRAS scores, and the finding of thinner mPFC in stressed subjects remained when adding the MADRAS scores as the covariate of no interest. Furthermore, the MADRAS scores showed no covariation with thickness of the mPFC. Third, the cortisol levels were normal in our stressed subjects, whereas they have been found to be high in a large portion of patients with genuine depression (Kasckow et al. 2001). We recently also found that women suffering from chronic occupational stress had an elevated reaction to allopregnanolone (Backstrom et al. 2013), which differs from the diminished allopregnanolone response that has been reported among depressed women (Girdler and Klatzkin 2007). Emotional reactions to chronic stress and major depression may, thus, represent separate constructs. They share, however, certain symptoms, perhaps due to the affection of similar limbic networks. Indeed, the higher MADRAS scores among the 6 stressed subjects who were not diagnosed as depressed could be an effect of this comorbidity. Given that depression can be triggered by stress, it is also possible that these subjects were developing depression, which was at a subclinical level at the time of investigation.

Methodological Issues

The methods used in this study were standard and have been used in several earlier studies yielding comparable results (Fischl et al. 2002; Sowell et al. 2007; Geuze et al. 2008; Lerch et al. 2008; Landre et al. 2010; Lyoo et al. 2011; Bruehl et al. 2013; Dickie et al. 2013). The hippocampal volumes were generally larger than when employing manual delineation (Blix et al. 2013) but were similar to those described in studies using the segmentation method of the FreeSurfer software (Doring et al. 2011; Morey et al. 2012). This systematic difference has been described previously (Tae et al. 2008; Doring et al. 2011) and may partly be attributed to the fact that FreeSurfer segments the entire hippocampus including the tail, which is not always included in the manually delineated hippocampal volumes.

The criteria for subject selection were conservative due to our ambition to optimize the possibility of investigating whether tentative cerebral changes were attributable to occupational stress. Because co-morbid stress factors (such as life traumas, depression, and PTSD) were excluded and the groups were matched for the education, and type of occupation, occupational stress was the major common denominator in the stressed group, and it seems reasonable that the data reliably reflected cerebral changes in relation to this specific factor.

Finally, it is worth mentioning that a more thorough presentation of the neurocognitive performance of these subjects on a detailed battery of neuropsychological tests will be presented in a separate manuscript.

Conclusion

In summary, the present findings indicate that chronic occupational stress is associated with morphological cerebral changes that are distributed in a differentiated manner along the limbic and paralimbic brain structures. Although needing longitudinal studies to be validated, the collective data provide reason to believe that the observed changes are more likely to represent effects of stress than causes of it. The changes correlated with the stress scores in a differentiated manner, and were located in regions known to mediate functions that were also shown to be impaired in our stressed subjects. Based on previous animal data and emerging studies of other stress conditions (PTSD), it is hypothesized that an excessive release of glutamate, perhaps in tandem with a dysregulation of HPA, leads to a cascade of events, ultimately leading to the detected changes—a development that hopefully is reversible. This condition needs to be considered as a stress illness, whose sufferers deserve proper and swift treatment.

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/.

Funding

The Swedish Council for Working Life and Social Research (FAS), the Swedish Research Council, and VINNOVA are acknowledged for their financial support.

Notes

Special acknowledgment is given to Dr Walter Ossika, who recruited patients, and Emilia Lentini and Elin Alzen, who preprocessed FreeSurfer data. Professor Hans Forsberg is acknowledged for his fruitful discussions. Dr Savic had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Conflict of Interest: None declared.

References

Ahola
K
Honkonen
T
Isometsa
E
Kalimo
R
Nykyri
E
Koskinen
S
Aromaa
A
Lonnqvist
J
.
2006
.
Burnout in the general population. Results from the Finnish Health 2000 Study
.
Soc Psychiatry Psychiatr Epidemiol
 .
41
:
11
17
.
Akerstedt
T
Knutsson
A
Westerholm
P
Theorell
T
Alfredsson
L
Kecklund
G
.
2002
.
Work organisation and unintentional sleep: results from the WOLF study
.
Occup Environl Med
 .
59
:
595
600
.
Arendt
DH
Smith
JP
Bastida
CC
Prasad
MS
Oliver
KD
Eyster
KM
Summers
TR
Delville
Y
Summers
CH
.
2012
.
Contrasting hippocampal and amygdalar expression of genes related to neural plasticity during escape from social aggression
.
Physiol Behav
 .
107
:
670
679
.
Arnsten
AF
.
2009
.
Stress signalling pathways that impair prefrontal cortex structure and function
.
Nat Rev Neurosci
 .
10
:
410
422
.
Asberg
M
Grape
T
Krakau
I
Nygren
A
Rohde
M
Wahlberg
A
Wahrborg
P
.
2010
.
[Stress as the cause of mental illness]
.
Lakartidningen
 .
107
:
1307
1310
.
Backstrom
T
Bixo
M
Nyberg
S
Savic
I
.
2013
.
Increased neurosteroid sensitivity—an explanation to symptoms associated with chronic work related stress in women?
Psychoneuroendocrinology
 .
38
:
1078
1089
.
Baker
LM
Williams
LM
Korgaonkar
MS
Cohen
RA
Heaps
JM
Paul
RH
.
2013
.
Impact of early vs. late childhood early life stress on brain morphometrics
.
Brain Imaging Behav
 .
7
:
196
203
.
Bengzon
J
Kokaia
Z
Elmer
E
Nanobashvili
A
Kokaia
M
Lindvall
O
.
1997
.
Apoptosis and proliferation of dentate gyrus neurons after single and intermittent limbic seizures
.
Proc Natl Acad Sci USA
 .
94
:
10432
10437
.
Bernal
F
Saura
J
Ojuel
J
Mahy
N
.
2000
.
Differential vulnerability of hippocampus, basal ganglia, and prefrontal cortex to long-term NMDA excitotoxicity
.
Exp Neurol
 .
161
:
686
695
.
Blix
E
Perski
A
Berglund
H
Savic
I
.
2013
.
Long-term occupational stress is associated with regional reductions in brain tissue volumes
.
PLoS One
 .
8
:
e64065
.
Bonne
O
Brandes
D
Gilboa
A
Gomori
JM
Shenton
ME
Pitman
RK
Shalev
AY
.
2001
.
Longitudinal MRI study of hippocampal volume in trauma survivors with PTSD
.
Am J Psychiatry
 .
158
:
1248
1251
.
Boyle
LM
.
2013
.
A neuroplasticity hypothesis of chronic stress in the basolateral amygdala
.
Yale J Biol Med
 .
86
:
117
125
.
Bremner
JD
.
2007
.
Neuroimaging in posttraumatic stress disorder and other stress-related disorders
.
Neuroimaging Clin N Am
 .
17
:
523
538
, ix
.
Bremner
JD
Elzinga
B
Schmahl
C
Vermetten
E
.
2008
.
Structural and functional plasticity of the human brain in posttraumatic stress disorder
.
Prog Brain Res
 .
167
:
171
186
.
Brown
ES
Zaidel
L
Allen
G
McColl
R
Vazquez
M
Ringe
WK
.
2010
.
Effects of lamotrigine on hippocampal activation in corticosteroid-treated patients
.
J Affect Disord
 .
126
:
415
419
.
Brown
SM
Henning
S
Wellman
CL
.
2005
.
Mild, short-term stress alters dendritic morphology in rat medial prefrontal cortex
.
Cereb Cortex
 .
15
:
1714
1722
.
Bruehl
H
Preissler
S
Heuser
I
Heekeren
HR
Roepke
S
Dziobek
I
.
2013
.
Increased prefrontal cortical thickness is associated with enhanced abilities to regulate emotions in PTSD-free women with borderline personality disorder
.
PLoS One
 .
8
:
e65584
.
Chao
L
Weiner
M
Neylan
T
.
2013
.
Regional cerebral volumes in veterans with current versus remitted posttraumatic stress disorder
.
Psychiatry Res
 .
213
:
193
201
.
Chen
Q
Harris
C
Brown
CS
Howe
A
Surmeier
DJ
Reiner
A
.
1995
.
Glutamate-mediated excitotoxic death of cultured striatal neurons is mediated by non-NMDA receptors
.
Exp Neurol
 .
136
:
212
224
.
Chida
Y
Steptoe
A
.
2009
.
Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis
.
Biol Psychol
 .
80
:
265
278
.
Ciumas
C
Wahlin
TB
Espino
C
Savic
I
.
2010
.
The dopamine system in idiopathic generalized epilepsies: identification of syndrome-related changes
.
Neuroimage
 .
51
:
606
615
.
Cohen
RA
Grieve
S
Hoth
KF
Paul
RH
Sweet
L
Tate
D
Gunstad
J
Stroud
L
McCaffery
J
Hitsman
B
et al
2006
.
Early life stress and morphometry of the adult anterior cingulate cortex and caudate nuclei
.
Biol Psychiatry
 .
59
:
975
982
.
Colberg
C
Antonow-Schlorke
I
Muller
T
Schubert
H
Witte
OW
Schwab
M
.
2004
.
Recovery of glucocorticoid-related loss of synaptic density in the fetal sheep brain at 0.75 of gestation
.
Neurosci Lett
 .
364
:
130
134
.
Conrad
CD
.
2008
.
Chronic stress-induced hippocampal vulnerability: the glucocorticoid vulnerability hypothesis
.
Rev Neurosci
 .
19
:
395
411
.
Conrad
CD
LeDoux
JE
Magarinos
AM
McEwen
BS
.
1999
.
Repeated restraint stress facilitates fear conditioning independently of causing hippocampal CA3 dendritic atrophy
.
Behav Neurosci
 .
113
:
902
913
.
Copertaro
A
Barbaresi
M
Tarsitani
L
Battisti
F
Baldassari
M
Picardi
A
Biondi
M
.
2007
.
[Fast stress evaluation in nurses]
.
G Ital Med Lav Ergon
 .
29
:
350
352
.
Copertaro
A
Bracci
M
Amati
M
Mocchegiani
E
Barbaresi
M
Copertaro
B
Santarelli
L
.
2010
.
[Biological risk and health care workers: analysis of the effects of work chronobiology on the immune system]
.
Med Lav
 .
101
:
427
436
.
De Bellis
MD
Hall
J
Boring
AM
Frustaci
K
Moritz
G
.
2001
.
A pilot longitudinal study of hippocampal volumes in pediatric maltreatment-related posttraumatic stress disorder
.
Biol Psychiatry
 .
50
:
305
309
.
Deykin
EY
Keane
TM
Kaloupek
D
Fincke
G
Rothendler
J
Siegfried
M
Creamer
K
.
2001
.
Posttraumatic stress disorder and the use of health services
.
Psychosom Med
 .
63
:
835
841
.
Dickie
EW
Brunet
A
Akerib
V
Armony
JL
.
2013
.
Anterior cingulate cortical thickness is a stable predictor of recovery from post-traumatic stress disorder
.
Psychol Med
 .
43
:
645
653
.
Doring
TM
Kubo
TT
Cruz
LC
Jr
Juruena
MF
Fainberg
J
Domingues
RC
Gasparetto
EL
.
2011
.
Evaluation of hippocampal volume based on MR imaging in patients with bipolar affective disorder applying manual and automatic segmentation techniques
.
J Magn Reson Imaging
 .
33
:
565
572
.
Egner
T
Etkin
A
Gale
S
Hirsch
J
.
2008
.
Dissociable neural systems resolve conflict from emotional versus nonemotional distracters
.
Cereb Cortex
 .
18
:
1475
1484
.
Fernandez Torres
B
Roldan Perez
LM
Guerra Velez
A
Roldan Rodriguez
T
Gutierrez Guillen
A
De las Mulas Bejar
M
.
2006
.
[Prevalence of burnout among anesthesiologists at Hospital Universitario Virgen Macarena de Sevilla]
.
Rev Esp Anestesiol Reanim
 .
53
:
359
362
.
Filipovic
BR
Djurovic
B
Marinkovic
S
Stijak
L
Aksic
M
Nikolic
V
Starcevic
A
Radonjic
V
.
2011
.
Volume changes of corpus striatum, thalamus, hippocampus and lateral ventricles in posttraumatic stress disorder (PTSD) patients suffering from headaches and without therapy
.
Cent Eur Neurosurg
 .
72
:
133
137
.
Fischl
B
Dale
AM
.
2000
.
Measuring the thickness of the human cerebral cortex from magnetic resonance images
.
Proc Natl Acad Sci USA
 .
97
:
11050
11055
.
Fischl
B
Salat
DH
Busa
E
Albert
M
Dieterich
M
Haselgrove
C
van der Kouwe
A
Killiany
R
Kennedy
D
Klaveness
S
et al
2002
.
Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain
.
Neuron
 .
33
:
341
355
.
Fischl
B
van der Kouwe
A
Destrieux
C
Halgren
E
Segonne
F
Salat
DH
Busa
E
Seidman
LJ
Goldstein
J
Kennedy
D
et al
2004
.
Automatically parcellating the human cerebral cortex
.
Cereb Cortex
 .
14
:
11
22
.
Fries
E
Dettenborn
L
Kirschbaum
C
.
2009
.
The cortisol awakening response (CAR): facts and future directions
.
Int J Psychophysiol
 .
72
:
67
73
.
Friess
E
Schmid
D
Modell
S
Brunner
H
Lauer
CJ
Holsboer
F
Ising
M
.
2008
.
Dex/CRH-test response and sleep in depressed patients and healthy controls with and without vulnerability for affective disorders
.
J Psychiatr Res
 .
42
:
1154
1162
.
Galbois
A
Rudler
M
Massard
J
Fulla
Y
Bennani
A
Bonnefont-Rousselot
D
Thibault
V
Reignier
S
Bourrier
A
Poynard
T
et al
2010
.
Assessment of adrenal function in cirrhotic patients: salivary cortisol should be preferred
.
J Hepatol
 .
52
:
839
845
.
Geuze
E
Westenberg
HG
Heinecke
A
de Kloet
CS
Goebel
R
Vermetten
E
.
2008
.
Thinner prefrontal cortex in veterans with posttraumatic stress disorder
.
Neuroimage
 .
41
:
675
681
.
Gilboa
A
Shalev
AY
Laor
L
Lester
H
Louzoun
Y
Chisin
R
Bonne
O
.
2004
.
Functional connectivity of the prefrontal cortex and the amygdala in posttraumatic stress disorder
.
Biol Psychiatry
 .
55
:
263
272
.
Girdler
SS
Klatzkin
R
.
2007
.
Neurosteroids in the context of stress: implications for depressive disorders
.
Pharmacol Ther
 .
116
:
125
139
.
Goldstein
JM
Jerram
M
Abbs
B
Whitfield-Gabrieli
S
Makris
N
.
2010
.
Sex differences in stress response circuitry activation dependent on female hormonal cycle
.
J Neurosci
 .
30
:
431
438
.
Grossi
G
Perski
A
Ekstedt
M
Johansson
T
Lindstrom
M
Holm
K
.
2005
.
The morning salivary cortisol response in burnout
.
J Psychosom Res
 .
59
:
103
111
.
Grossi
G
Perski
A
Evengard
B
Blomkvist
V
Orth-Gomer
K
.
2003
.
Physiological correlates of burnout among women
.
J Psychosom Res
 .
55
:
309
316
.
Gurvits
TV
Shenton
ME
Hokama
H
Ohta
H
Lasko
NB
Gilbertson
MW
Orr
SP
Kikinis
R
Jolesz
FA
McCarley
RW
et al
1996
.
Magnetic resonance imaging study of hippocampal volume in chronic, combat-related posttraumatic stress disorder
.
Biol Psychiatry
 .
40
:
1091
1099
.
He
Y
Chen
Z
Evans
A
.
2008
.
Structural insights into aberrant topological patterns of large-scale cortical networks in Alzheimer's disease
.
J Neurosci
 .
28
:
4756
4766
.
Holmes
TH
Rahe
RH
.
1967
.
The Social Readjustment Rating Scale
.
J Psychosom Res
 .
11
:
213
218
.
Hunter
RG
Bellani
R
Bloss
E
Costa
A
McCarthy
K
McEwen
BS
.
2009
.
Regulation of kainate receptor subunit mRNA by stress and corticosteroids in the rat hippocampus
.
PloS One
 .
4
:
e4328
.
Iijima
M
Shimazaki
T
Ito
A
Chaki
S
.
2007
.
Effects of metabotropic glutamate 2/3 receptor antagonists in the stress-induced hyperthermia test in singly housed mice
.
Psychopharmacology
 .
190
:
233
239
.
Isovich
E
Mijnster
MJ
Flugge
G
Fuchs
E
.
2000
.
Chronic psychosocial stress reduces the density of dopamine transporters
.
Eur J Neurosci
 .
12
:
1071
1078
.
Jovanovic
H
Perski
A
Berglund
H
Savic
I
.
2011
.
Chronic stress is linked to 5-HT(1A) receptor changes and functional disintegration of the limbic networks
.
Neuroimage
 .
55
:
1178
1188
.
Karlamangla
AS
Friedman
EM
Seeman
TE
Stawksi
RS
Almeida
DM
.
2013
.
Daytime trajectories of cortisol: demographic and socioeconomic differences—findings from the National Study of Daily Experiences
.
Psychoneuroendocrinology
 .
38
:
2585
2597
.
Kasai
K
Yamasue
H
Gilbertson
MW
Shenton
ME
Rauch
SL
Pitman
RK
.
2008
.
Evidence for acquired pregenual anterior cingulate gray matter loss from a twin study of combat-related posttraumatic stress disorder
.
Biol Psychiatry
 .
63
:
550
556
.
Kasckow
JW
Baker
D
Geracioti
TD
Jr
.
2001
.
Corticotropin-releasing hormone in depression and post-traumatic stress disorder
.
Peptides
 .
22
:
845
851
.
Koenigs
M
Grafman
J
.
2009
.
Posttraumatic stress disorder: the role of medial prefrontal cortex and amygdala
.
Neuroscientist
 .
15
:
540
548
.
Kruggel
F
Bruckner
MK
Arendt
T
Wiggins
CJ
von Cramon
DY
.
2003
.
Analyzing the neocortical fine-structure
.
Med Image Anal
 .
7
:
251
264
.
Kuo
JR
Kaloupek
DG
Woodward
SH
.
2012
.
Amygdala volume in combat-exposed veterans with and without posttraumatic stress disorder: a cross-sectional study
.
Arch Gen Psychiatry
 .
69
:
1080
1086
.
Ladefoged
O
Hougaard
KS
Hass
U
Sorensen
IK
Lund
SP
Svendsen
GW
Lam
HR
.
2004
.
Effects of combined prenatal stress and toluene exposure on apoptotic neurodegeneration in cerebellum and hippocampus of rats
.
Basic Clin Pharmacol Toxicol
 .
94
:
169
176
.
Landre
L
Destrieux
C
Baudry
M
Barantin
L
Cottier
JP
Martineau
J
Hommet
C
Isingrini
M
Belzung
C
Gaillard
P
et al
2010
.
Preserved subcortical volumes and cortical thickness in women with sexual abuse-related PTSD
.
Psychiatry Res
 .
183
:
181
186
.
Langelaan
S
Bakker
AB
Schaufeli
WB
van Rhenen
W
van Doornen
LJ
.
2007
.
Is burnout related to allostatic load?
Int J Behav Med
 .
14
:
213
221
.
Lazarus
RS
Folkman
S
.
1987
.
Transactional theory and research on emotions and coping
.
Eur J Pers
 .
1
:
141
170
.
Lerch
JP
Pruessner
J
Zijdenbos
AP
Collins
DL
Teipel
SJ
Hampel
H
Evans
AC
.
2008
.
Automated cortical thickness measurements from MRI can accurately separate Alzheimer's patients from normal elderly controls
.
Neurobiol Aging
 .
29
:
23
30
.
Leuner
B
Shors
TJ
.
2012
.
Stress, anxiety, and dendritic spines: what are the connections?
Neuroscience
 .
251
:
108
119
.
Lindauer
RJ
Vlieger
EJ
Jalink
M
Olff
M
Carlier
IV
Majoie
CB
den Heeten
GJ
Gersons
BP
.
2004
.
Smaller hippocampal volume in Dutch police officers with posttraumatic stress disorder
.
Biol Psychiatry
 .
56
:
356
363
.
Liston
C
McEwen
BS
Casey
BJ
.
2009
.
Psychosocial stress reversibly disrupts prefrontal processing and attentional control
.
Proc Natl Acad Sci USA
 .
106
:
912
917
.
Lowy
MT
Gault
L
Yamamoto
BK
.
1993
.
Adrenalectomy attenuates stress-induced elevations in extracellular glutamate concentrations in the hippocampus
.
J Neurochem
 .
61
:
1957
1960
.
Lupien
SJ
Parent
S
Evans
AC
Tremblay
RE
Zelazo
PD
Corbo
V
Pruessner
JC
Seguin
JR
.
2011
.
Larger amygdala but no change in hippocampal volume in 10-year-old children exposed to maternal depressive symptomatology since birth
.
Proc Natl Acad Sci USA
 .
108
:
14324
14329
.
Lyoo
IK
Kim
JE
Yoon
SJ
Hwang
J
Bae
S
Kim
DJ
.
2011
.
The neurobiological role of the dorsolateral prefrontal cortex in recovery from trauma. Longitudinal brain imaging study among survivors of the South Korean subway disaster
.
Arch Gen Psychiatry
 .
68
:
701
713
.
Magarinos
AM
McEwen
BS
.
1995
.
Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: involvement of glucocorticoid secretion and excitatory amino acid receptors
.
Neuroscience
 .
69
:
89
98
.
Matsukura
TS
Marturano
EM
Oishi
J
.
2002
.
[The Saranson's Social Support questionnaire: studies regarding its adaptation to portuguese]
.
Rev Lat-Am Enfermagem
 
10
:
675
681
.
McEwen
BS
.
1994
.
Corticosteroids and hippocampal plasticity
.
Ann N Y Acad Sci
 
746
:
134
142
;
discussion
142
134
,
178
139
.
McEwen
BS
.
2000a
.
The neurobiology of stress: from serendipity to clinical relevance
.
Brain Res
 .
886
:
172
189
.
McEwen
BS
.
2000b
.
Protective and damaging effects of stress mediators: central role of the brain
.
Prog Brain Res
 .
122
:
25
34
.
McEwen
BS
Conrad
CD
Kuroda
Y
Frankfurt
M
Magarinos
AM
McKittrick
C
.
1997
.
Prevention of stress-induced morphological and cognitive consequences
.
Eur Neuropsychopharmacol
 .
7
(Suppl 3)
:
S323
S328
.
Melamed
S
Ugarten
U
Shirom
A
Kahana
L
Lerman
Y
Froom
P
.
1999
.
Chronic burnout, somatic arousal and elevated salivary cortisol levels
.
J Psychosom Res
 .
46
:
591
598
.
Moch
SL
Panz
VR
Joffe
BI
Havlik
I
Moch
JD
.
2003
.
Longitudinal changes in pituitary-adrenal hormones in South African women with burnout
.
Endocrine
 .
21
:
267
272
.
Mommersteeg
PM
Heijnen
CJ
Verbraak
MJ
van Doornen
LJ
.
2006
.
Clinical burnout is not reflected in the cortisol awakening response, the day-curve or the response to a low-dose dexamethasone suppression test
.
Psychoneuroendocrinology
 .
31
:
216
225
.
Mommersteeg
PM
Keijsers
GP
Heijnen
CJ
Verbraak
MJ
van Doornen
LJ
.
2006
.
Cortisol deviations in people with burnout before and after psychotherapy: a pilot study
.
Health Psychol
 .
25
:
243
248
.
Montgomery
S
Asberg
M
Traskman
L
Montgomery
D
.
1978
.
Cross cultural studies on the use of CPRS in English and Swedish depressed patients
.
Acta Psychiatr Scand Suppl
 .
(271)
:
33
37
.
Morey
RA
Gold
AL
LaBar
KS
Beall
SK
Brown
VM
Haswell
CC
Nasser
JD
Wagner
HR
McCarthy
G
Mid-Atlantic
MW
.
2012
.
Amygdala volume changes in posttraumatic stress disorder in a large case-controlled veterans group
.
Arch Gen Psychiatry
 .
69
:
1169
1178
.
Oldfield
RC
.
1971
.
The assessment and analysis of handedness: the Edinburgh inventory
.
Neuropsychologia
 .
9
:
97
113
.
Papagni
SA
Benetti
S
Arulanantham
S
McCrory
E
McGuire
P
Mechelli
A
.
2011
.
Effects of stressfull life events on human brain structure: a longitudinal voxel-based morphometry study
.
Stress
 .
14
:
227
232
.
Pavlisa
G
Papa
J
Pavic
L
.
2006
.
Bilateral MR volumetry of the amygdala in chronic PTSD patients
.
Coll Antropol
 .
30
:
565
568
.
Pavlisa
G
Papa
J
Pavic
L
Pavlisa
G
.
2006
.
Bilateral MR volumetry of the amygdala in chronic PTSD patients
.
Coll Antropol
 .
30
:
565
568
.
Petrides
M
Pandya
DN
.
1988
.
Association fiber pathways to the frontal cortex from the superior temporal region in the rhesus monkey
.
J Comp Neurol
 .
273
:
52
66
.
Petrides
M
Pandya
DN
.
1999
.
Dorsolateral prefrontal cortex: comparative cytoarchitectonic analysis in the human and the macaque brain and corticocortical connection patterns
.
Eur J Neurosci
 .
11
:
1011
1036
.
Pezawas
L
Verchinski
BA
Mattay
VS
Callicott
JH
Kolachana
BS
Straub
RE
Egan
MF
Meyer-Lindenberg
A
Weinberger
DR
.
2004
.
The brain-derived neurotrophic factor val66met polymorphism and variation in human cortical morphology
.
J Neurosci
 .
24
:
10099
10102
.
Pitman
RK
Rasmusson
AM
Koenen
KC
Shin
LM
Orr
SP
Gilbertson
MW
Milad
MR
Liberzon
I
.
2012
.
Biological studies of post-traumatic stress disorder
.
Nat Rev Neurosci
 .
13
:
769
787
.
Porcelli
AJ
Cruz
D
Wenberg
K
Patterson
MD
Biswal
BB
Rypma
B
.
2008
.
The effects of acute stress on human prefrontal working memory systems
.
Physiol Behav
 .
95
:
282
289
.
Pruessner
JC
Hellhammer
DH
Kirschbaum
C
.
1999
.
Burnout, perceived stress, and cortisol responses to awakening
.
Psychosom Med
 .
61
:
197
204
.
Radley
JJ
Rocher
AB
Rodriguez
A
Ehlenberger
DB
Dammann
M
McEwen
BS
Morrison
JH
Wearne
SL
Hof
PR
.
2008
.
Repeated stress alters dendritic spine morphology in the rat medial prefrontal cortex
.
J Comp Neurol
 .
507
:
1141
1150
.
Roozendaal
B
McReynolds
JR
McGaugh
JL
.
2004
.
The basolateral amygdala interacts with the medial prefrontal cortex in regulating glucocorticoid effects on working memory impairment
.
J Neurosci
 .
24
:
1385
1392
.
Rydmark
I
Wahlberg
K
Ghatan
PH
Modell
S
Nygren
A
Ingvar
M
Asberg
M
Heilig
M
.
2006
.
Neuroendocrine, cognitive and structural imaging characteristics of women on longterm sickleave with job stress-induced depression
.
Biol Psychiatry
 .
60
:
867
873
.
Ryzhavskii
BY
Vasil'eva
EV
Sokolova
TV
.
2003
.
Morphology of the cerebellum in the progeny of female rats exposed to long-term emotional stress before pregnancy
.
Bull Exp Biol Med
 .
135
:
206
208
.
Sandstrom
A
Rhodin
IN
Lundberg
M
Olsson
T
Nyberg
L
.
2005
.
Impaired cognitive performance in patients with chronic burnout syndrome
.
Biol Psychol
 .
69
:
271
279
.
Sapolsky
RM
.
2000
.
Stress hormones: good and bad
.
Neurobiol Dis
 .
7
:
540
542
.
Schaufeli
WB
Van Dierendonck
D
.
1995
.
A cautionary note about the cross-national and clinical validity of cut-off points for the Maslach Burnout Inventory
.
Psychol Rep
 .
76
:
1083
1090
.
Shors
TJ
Chua
C
Falduto
J
.
2001
.
Sex differences and opposite effects of stress on dendritic spine density in the male versus female hippocampus
.
J Neurosci
 .
21
:
6292
6297
.
Sowell
ER
Peterson
BS
Kan
E
Woods
RP
Yoshii
J
Bansal
R
Xu
D
Zhu
H
Thompson
PM
Toga
AW
.
2007
.
Sex differences in cortical thickness mapped in 176 healthy individuals between 7 and 87 years of age
.
Cereb Cortex
 .
17
:
1550
1560
.
Sowell
ER
Peterson
BS
Thompson
PM
Welcome
SE
Henkenius
AL
Toga
AW
.
2003
.
Mapping cortical change across the human life span
.
Nat Neurosci
 .
6
:
309
315
.
Stenlund
T
Ahlgren
C
Lindahl
B
Burell
G
Knutsson
A
Stegmayr
B
Birgander
LS
.
2007
.
Patients with burnout in relation to gender and a general population
.
Scand J Public Health
 .
35
:
516
523
.
Tae
WS
Kim
SS
Lee
KU
Nam
EC
Kim
KW
.
2008
.
Validation of hippocampal volumes measured using a manual method and two automated methods (FreeSurfer and IBASPM) in chronic major depressive disorder
.
Neuroradiology
 .
50
:
569
581
.
Ulrich-Lai
YM
Herman
JP
.
2009
.
Neural regulation of endocrine and autonomic stress responses
.
Nat Rev Neurosci
 .
10
:
397
409
.
van Harmelen
AL
van Tol
MJ
van der Wee
NJ
Veltman
DJ
Aleman
A
Spinhoven
P
van Buchem
MA
Zitman
FG
Penninx
BW
Elzinga
BM
.
2010
.
Reduced medial prefrontal cortex volume in adults reporting childhood emotional maltreatment
.
Biol Psychiatry
 .
68
:
832
838
.
Villarreal
G
Hamilton
DA
Petropoulos
H
Driscoll
I
Rowland
LM
Griego
JA
Kodituwakku
PW
Hart
BL
Escalona
R
Brooks
WM
.
2002
.
Reduced hippocampal volume and total white matter volume in posttraumatic stress disorder
.
Biol Psychiatry
 .
52
:
119
125
.
Wahlberg
K
Ghatan
PH
Modell
S
Nygren
A
Ingvar
M
Asberg
M
Heilig
M
.
2009
.
Suppressed neuroendocrine stress response in depressed women on job-stress-related long-term sick leave: a stable marker potentially suggestive of preexisting vulnerability
.
Biol Psychiatry
 .
65
:
742
747
.
Wang
HL
Tsai
LY
Lee
EH
.
2000
.
Corticotropin-releasing factor produces a protein synthesis—dependent long-lasting potentiation in dentate gyrus neurons
.
J Neurophysiol
 .
83
:
343
349
.
Wang
J
Korczykowski
M
Rao
H
Fan
Y
Pluta
J
Gur
RC
McEwen
BS
Detre
JA
.
2007
.
Gender difference in neural response to psychological stress
.
Soc Cogn Affect Neurosci
 .
2
:
227
239
.
Weber
M
Killgore
WD
Rosso
IM
Britton
JC
Schwab
ZJ
Weiner
MR
Simon
NM
Pollack
MH
Rauch
SL
.
2013
.
Voxel-based morphometric gray matter correlates of posttraumatic stress disorder
.
J Anxiety Disord
 .
27
:
413
419
.
Wellman
CL
.
2001
.
Dendritic reorganization in pyramidal neurons in medial prefrontal cortex after chronic corticosterone administration
.
J Neurobiol
 .
49
:
245
253
.
Wong
EY
Herbert
J
.
2004
.
The corticoid environment: a determining factor for neural progenitors’ survival in the adult hippocampus
.
Eur J Neurosci
 .
20
:
2491
2498
.
Woon
FL
Hedges
DW
.
2009
.
Amygdala volume in adults with posttraumatic stress disorder: a meta-analysis
.
J Neuropsychiatry Clin Neurosci
 .
21
:
5
12
.
Yin
Y
Jin
C
Hu
X
Duan
L
Li
Z
Song
M
Chen
H
Feng
B
Jiang
T
Jin
H
et al
2011
.
Altered resting-state functional connectivity of thalamus in earthquake-induced posttraumatic stress disorder: a functional magnetic resonance imaging study
.
Brain Res
 .
1411
:
98
107
.
Zhang
J
Tan
Q
Yin
H
Zhang
X
Huan
Y
Tang
L
Wang
H
Xu
J
Li
L
.
2011
.
Decreased gray matter volume in the left hippocampus and bilateral calcarine cortex in coal mine flood disaster survivors with recent onset PTSD
.
Psychiatry Res
 .
192
:
84
90
.