Spontaneous alterations of regional brain activity in patients with adult generalized anxiety disorder.

OBJECTIVE: We aimed to examine how spontaneous brain activity might be related to the pathophysiology of generalized anxiety disorder (GAD). PATIENTS AND METHODS: Using resting-state functional MRI, we examined spontaneous regional brain activity in 31 GAD patients (mean age, 36.87±9.16 years) and 36 healthy control participants (mean age, 39.53±8.83 years) matched for age, education, and sex from December 2014 to October 2015. We performed a two-sample t-test on the voxel-based analysis of the regional homogeneity (ReHo) maps. We used Pearson correlation analysis to compare scores from the Hamilton Anxiety Rating Scale, Hamilton Depression Rating Scale, State-Trait Anxiety Scale-Trait Scale, and mean ReHo values.
RESULTS: We found abnormal spontaneous activity in multiple regions of brain in GAD patients, especially in the sensorimotor cortex and emotional regions. GAD patients showed decreased ReHo values in the right orbital middle frontal gyrus, left anterior cingulate cortex, right middle frontal gyrus, and bilateral supplementary motor areas, with increased ReHo values in the left middle temporal gyrus, left superior temporal gyrus, and right superior occipital gyrus. The ReHo value of the left middle temporal gyrus correlated positively with the Hamilton Anxiety Rating Scale scores.
CONCLUSION: These results suggest that altered local synchronization of spontaneous brain activity may be related to the pathophysiology of GAD.

Introduction

Generalized anxiety disorder (GAD) is marked by excessive and uncontrollable anxiety and worry. GAD is a common anxiety disorder that shows a lifetime prevalence of 4.3%–5.9% and a 12-month prevalence of 0.2%–4.3%.1,2 GAD patients have a higher prevalence of lifetime suicidal thoughts and attempts than patients with other anxiety disorders. Currently, diagnosis of GAD is based mainly on clinical psychiatric signs and symptoms. However, such an approach might lead to misdiagnosis as a result of high rates of comorbidity with depression and agoraphobia.3 Therefore, an objective and stable method to improve the diagnostic accuracy of GAD is needed.

In recent years, researchers have used a variety of neuroimaging techniques to explore the pathogenesis of GAD. Using single-photon emission computed tomography, Lee et al4 found that the striatal dopamine transporter level in patients with GAD could have a role in the pathophysiology of GAD. Using proton magnetic resonance spectroscopy, Moon et al5 found that the choline/N-acetylaspartate metabolic changes in the dorsolateral prefrontal cortex in patients with GAD were closely related to symptom severity and cognitive dysfunction. In addition, diffusion tensor imaging studies found that adult patients with GAD showed lower fractional anisotropy in the uncinate fasciculus, which connects the amygdala and the frontal cortex.6,7

More researchers have focused on functional magnetic resonance imaging (fMRI) technology. Evidence from various task-based fMRI studies has indicated that an alteration occurs in the activation of the anterior cingulate cortex (ACC), prefrontal cortex (PFC), insula, amygdala, and other limbic brain regions when patients with GAD perform specific tasks.8–11 Differentiation of these functional alterations might reveal abnormal neural circuit and emotion regulation in GAD patients. However, task-based fMRI requires good cooperation from subjects and can only reflect the brain neuron activities associated with the specific tasks, implying that task-based fMRI might have less clinical practicability. Therefore, to understand how the brain operates in GAD, we must consider the spontaneous neuronal activity which consumes most of the brain’s energy.

The noninvasive resting-state fMRI (rs-fMRI), unlike the complex task-based fMRI, can be performed conveniently in clinics and is widely used to evaluate spontaneous neuronal activity. A systematic review of the literature showed that an increasing number of researchers have used rs-fMRI techniques and functional connectivity analysis to study other types of anxiety disorders including post-traumatic stress disorder, obsessive–compulsive disorder, social anxiety disorder, and panic disorder.12 However, regional spontaneous activity in GAD based on rs-fMRI techniques has not been extensively reported in previous studies. Moreover, the most commonly used is the method of functional connectivity in GAD based on rs-fMRI.13–15 Makovac et al14 showed reduction in connectivity between right amygdala and ventromedial PFC, enhanced coupling between left amygdala and ventral tegmental area, and increased connectivity between right amygdala and thalamus. Results highlighted amygdala functional connectivity as a longitudinal biomarker of symptom changes in GAD. Functional connectivity showed the interaction between brain regions.

Regional homogeneity (ReHo) is widely used with rs-fMRI to evaluate regional spontaneous activity across the whole brain. ReHo can measure the similarity of voxel of the time series of a given voxel with its nearest neighbors by calculating Kendall’s coefficient of concordance (KCC). ReHo has been applied commonly to various neuropsychiatric disorders, such as sleep disorders,16 transient ischemic attack,17 rolandic epilepsy,18 and depression.19 To our knowledge, the ReHo approach has not been used to investigate spontaneous brain activity in GAD.

In this study, we used the ReHo method based on rs-fMRI to gain additional insights into aberrant spontaneous neural activity in GAD patients. In accordance with previous findings, we hypothesized that compared with healthy controls, patients with GAD would exhibit abnormal ReHo activity, mainly in sensorimotor cortex and emotional regions. Moreover, we speculated that the ReHo values extracted from the abnormal regions would be correlated with an emotional scale, thereby providing potential imaging evidence to explain the clinical manifestations of GAD.

Patients and methods

Subjects

This study included a group of 31 GAD patients (mean age, 36.87±9.16 years; 15 males and 16 females). The study also included 36 healthy sex-, hand-, age-, and education-matched controls (mean age, 39.53±8.83 years; 13 males and 23 females). The study was approved by the Research Ethics Committee of People’s Hospital of Yuxi City (Yuxi, People’s Republic of China). All participants signed written informed consent agreement, and participants were reimbursed $50 for their time/efforts.

GAD patients were recruited from the Department of Rehabilitation Medicine at People’s Hospital of Yuxi City from December 2014 to October 2015. Diagnosis of GAD was determined by consensus of two experienced psychiatrists. The inclusion criteria for patients with GAD were as follows: 1) conforming to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Version IV; 2) had been complaining of uncontrollable anxiety and worry about everyday events and problems for at least 6 months; 3) had not taken antianxiety or antipsychotic drugs for at least 1 month before the study; 4) had a Hamilton Anxiety Rating Scale (HAMA) score ≥14; and 5) being younger than 60 years old.

The exclusion criteria for patients with GAD were as follows: 1) having other psychiatric disorders and secondary anxiety disorder; 2) exhibiting drug or alcohol dependence; 3) having had a past history of depression or episodes of depression; 4) having a 24-item Hamilton Depression Rating Scale (HAMD) score >20; 5) having been diagnosed with other significant medical conditions, such as diabetes or high blood pressure; 6) having an abnormal signal as verified by conventional T1 or T2 fluid-attenuated inversion recovery MRI; 7) being pregnant or breastfeeding. We excluded four GAD patients with abnormal signals in the conventional T2 image.

The inclusion criteria for healthy controls were as follows: 1) having a HAMA score <7 and a 24-item HAMD score <8; 2) a conventional MRI showing no organic disease; 3) had no history of inborn or other psychiatric disorder; and 4) had no metal foreign objects in his/her body.

Additional clinical measures

Before MRI examinations, the emotional quantifiable indicators of every subject were rated by the HAMA and the 24-item HAMD scales as well as the State–Trait Anxiety Inventory (STAI) scale.

MRI data acquisition

All MRI datasets were acquired using a 3 T MRI scanner with an eight-channel head coil (Ingenia; Philips, the Netherlands) in the MRI Department of People’s Hospital of Yuxi City. Each subject was supine with headphones to reduce noise and was positioned with foam pads to secure the head and reduce head motion. During the rs-fMRI data acquisition, all subjects were asked to keep their eyes closed, relax, hold still, not think of anything systematically, and not fall asleep. The rs-fMRI scans were performed using a gradient echo planar imaging sequence. The acquisition parameters were as follows: repeat time =2,000 ms; echo time =35 ms, flip angle =90°; field of view =230×230 mm; matrix =64×64; total volume =240; voxel size: 3.6×3.6×3.6 mm; slice thickness =3.6 mm with a 0.7 mm gap; 35 transverse plane parallel AC–PC line; 240 dynamic scanning; 8,400 images. The fMRI scanning for each subject lasted 8 minutes. After the MRI scanning, the patient was interviewed to assess whether he or she had complied with the instructions and was awake throughout the scan. Subjects who failed to comply with the instructions were excluded from the study.

Data processing and ReHo analyses

Data preprocessing was based on a Matlab 2012b operation interface, using the Data Processing Assistant for Resting-State fMRI integrated software (http://restfmri.net/forum/DPARSF).20 For data processing, the first 10 time points were discarded due to transient signal changes before magnetization reached a steady state and/or adaptation of the subject to the circumstances, leaving 230 time points for the preprocessing steps of slice timing. No subject’s data were discarded as a result of excessive head motion (>1.5 mm in translation or 1.5° in rotation).

Next, we compared the motion courses of the two groups using a two-sample t-test. This test confirmed no statistical significance. After head motion correction, the fMRI images were normalized to the Montreal Neurological Institute template (resampling voxel size =3×3×3 mm3). Linear trends and temporal filtering (band pass, 0.01–0.08 Hz) were removed to reduce very low frequency drift and physiologic high-frequency respiratory and cardiac noise.21

The ReHo calculation procedure used the Resting-State fMRI Data Analysis Toolkit (REST; http://resting-fmri.sourceforge.net), the same as described in previous studies.22 A KCC value (ReHo value) was used to measure the similarity of the time series of a given voxel to its nearest 26 neighbors.23 The KCC value calculation formula was as follows, where W is the KCC value of that given voxel, ranging from 0 to 1:

When a given cluster and its neighboring cluster in a time sequence are more consistent, the KCC value is close to 1. R is the sum rank of the ith time point. R– = [(n + 1) K]/2 is the mean of the Ris. K is the number of voxels within a measured cluster (here, K =27, the number of nearest neighboring 26 voxels plus that one given voxel), and n is the number for the ranks (here, n=150).

Normalized ReHo value is the KCC value of each voxel/the mean of the whole brain KCC. The individual ReHo map was obtained for each data set. Finally, we applied an 8 mm full width at half maximum spatial smoothing Gaussian kernel to reduce the spatial noise.

Statistical analysis

We used SPSS Version 20.0 (IBM Corporation, Armonk, NY, USA) for analyzing demographic data and clinical characteristics. A chi-square test was used for sex comparison. A two-sample t-test was performed to evaluate the differences in age, education, HAMA, HAMD, and State-Trait Anxiety Scale-Trait Scale (STAI-T) scores between patients with GAD and healthy controls, with the significance level set at P<0.05.

A two-sample t-test was also applied to compare the smReHo maps between GAD patients and healthy controls. Multiple comparison correction was performed by using the AlphaSim program in REST and a P-value <0.05. For each region of interest, the mean ReHo values for all voxels in the significant areas were extracted separately using the REST. Finally, Pearson correlation analysis was also adopted to clarify the relationship between the mean ReHo values in significantly different areas and the HAMA, HAMD-24, and STAI-T scores in SPSS 20.0 (P<0.05).

Results

Demographic and clinical variables

The study results demonstrated no significant differences in sex (P=0.310), age (P=0.232), and education (P=0.088) between the GAD group and the healthy control group. The HAMA, HAMD-24, and STAI-T scores of the GAD subjects were significantly higher than those of the healthy controls. The detailed demographic and clinical data are presented in Table 1.

Table 1

Demographic and clinical characteristics of GAD patients and HC

Characteristics	GAD(n=31)	HC(n=36)	t-value	P-value
Sex (male/female)	15/16	13/23	NA	0.3100
Age (years)	36.9±9.2	39.5±8.8	−1.2070	0.2320
Illness duration (years)	2.8±3.8	NA	NA	NA
Education (years)	10.5±3.9	11.9±3.0	−1.7350	0.0880
HAMA	22.3±5.3	1.1±1.9	21.1380	0.0000
HAMD-24	14.9±3.3	0.4±1.2	23.3980	0.0000
STAI-T	48.9±9.1	21.2±2.8	16.2870	0.0000

Note: Data are reported as mean ± SD.

Abbreviations: GAD, generalized anxiety disorder; HAMA, Hamilton Anxiety Rating Scale; HAMD-24, Hamilton Depression Rating Scale-24 items; HC, healthy controls; NA, not available; STAI-T, State–Trait Anxiety Scale-Trait Scale.

Alterations in ReHo values of GAD patients

In this study, GAD patients showed significantly different ReHo values in widespread areas when compared with healthy controls (voxel P<0.05, cluster P<0.05 with AlphaSim corrected). The detailed between-group differences of ReHo values are presented in Table 2 and Figures 1 and 2. Compared to the healthy controls, GAD patients had significantly decreased ReHo values in the right orbital middle frontal gyrus (MFG), left ACC, right MFG, and bilateral supplementary motor area (SMA). GAD patients showed increased ReHo values in the left middle temporal gyrus (MTG), left superior temporal gyrus (STG), and right superior occipital gyrus (SOG).

Table 2

Brain regions showing abnormal ReHo in GAD patients in comparison to HC

Brain regions	Voxels	MNI coordinates (mm)			Peak t-value
		x	y	z
Increased ReHo (GAD > HC)
Left middle temporal gyrus	57	−54	−3	−24	3.7628
Left superior temporal gyrus	99	−51	6	0	3.4414
Right superior occipital gyrus	67	30	−81	42	3.0177
Decreased ReHo (GAD < HC)
Right orbital middle frontal gyrus	82	12	60	−9	−3.6965
Left anterior cingulate cortex	56	−9	48	12	−3.2629
Right middle frontal gyrus	87	39	42	15	−4.1522
Right supplementary motor area	76	9	21	45	−3.5626
Left supplementary motor area	87	−3	24	48	−3.2133

Abbreviations: GAD, generalized anxiety disorder; HC, healthy controls; MNI, Montreal Neurological Institute; ReHo, regional homogeneity.

Figure 1

Brain regions showing abnormal ReHo in GAD patients in comparison to healthy controls.

Notes: The warm colors (positive value) represent increased ReHo areas and the cold colors (negative value) represent decreased ReHo areas. A T-score bar is shown on the right. Compared to the healthy controls, GAD patients had significantly decreased ReHo in the right orbital middle frontal gyrus, left anterior cingulate cortex, right middle frontal gyrus, and bilateral supplementary motor area. GAD patients demonstrated increased ReHo values in the left middle temporal gyrus, left superior temporal gyrus, and right superior occipital gyrus.

Abbreviations: GAD, generalized anxiety disorders; ReHo, regional homogeneity.

Figure 2

Bars of the mean ReHo values for the identified brain regions.

Note: All regions showed significant differences between GAD patients and healthy controls.

Abbreviations: ACC, anterior cingulate cortex; GAD, generalized anxiety disorders; HC, healthy controls; L, left; MFG, middle frontal gyrus; MTG, middle temporal gyrus; R, right; ReHo, regional homogeneity; SMA, supplementary motor area; SOG, superior occipital gyrus; STG, superior temporal gyrus.

The correlation between ReHo values and clinical variables in GAD patients

As shown in Figure 3A and B, the ReHo value in the left MTG showed significantly positive correlation with HAMA scores: r=0.488, P=0.005. In this study, no other significant correlations were found between the ReHo values in regions showing significant differences and other clinical variables.

Figure 3

Correlation between MTG-L and HAMA scores.

Note: (A, B) The ReHo value of the MTG-L correlated positively with HAMA scores.

Abbreviations: HAMA, Hamilton Anxiety Rating Scale; L, left; MTG, middle temporal gyrus; ReHo, regional homogeneity.

Discussion

Our study used ReHo to investigate the synchrony of regional spontaneous activity in rs-fMRI among patients with GAD. We found that patients with GAD showed significantly lower ReHo in the right orbital MFG, left ACC, right MFG, and bilateral SMA, as well as higher ReHo in the left MTG, left STG, and right SOG. The ReHo value of the left MTG had a significant positive correlation with the HAMA score.

The decreased ReHo values in the right orbital MFG and right MFG in the GAD group reflect destruction of local synchronization of spontaneous low-frequency fluctuations dependent on blood oxygenation levels in these two regions. This finding implies functional deficits. Frontal lobe dysfunction is characterized clinically by decrease of spontaneity, initiative, insight, judgment, abstraction, perseverance, and response inhibition. Previous neuroimaging studies showed that PFC played an important role in anxiety. Kim et al24 found a variety of anxiety disorders were typically characterized by hyperactivity of the amygdala and hypoactivity of the PFC. The disrupted emotional and cognitive processes were thought to be a crucial component of symptomatology in the pathology of anxiety.

In addition, previous studies found abnormal activation in the MFG of GAD patients. Blair et al25 reported that GAD patients had significantly increased response to angry expressions in a lateral region of the MFG, as compared to healthy individuals. This finding also related to self-reported anxiety in patients with GAD. Using 99mTc-hexamethylpropyleneamine oxime single-photon emission computed tomography, Kalk et al26 found increased perfusion in the right MFG in untreated GAD patients. Together with previous neuroimaging studies related to GAD, our study suggested that functional disturbance in the PFC might be a direct neural mechanism of GAD.

In addition, GAD patients showed decreased ReHo in the left ACC. The ACC is a critical component of the limbic system that has been confirmed as having a widely structural and functional connection in the PFC.27 The ACC plays an important role in emotion regulation and autonomic control,28 and the ventral ACC is involved in the elimination and adjustment of negative emotion.29 Previous fMRI studies have shown that the ACC is activated while GAD patients perform the emotional arousal task30 and it decreases while the patients perform the conflict mission and emotional face recognition tasks.31,32 Moreover, a systematic review showed GAD has been linked to hypofunction in the PFC and ACC that is associated with emotion regulation.33 The emotion dysregulation theory showed that patients with GAD experience emotional hyperarousal, contributing to maladaptive emotion regulation and unsuccessful attempts to either minimize or over-control emotions.34 Consequently, in our study, the decreased ReHo in the ACC suggests dysfunction of spontaneous neural activity, which may relate to the emotional disturbance of uncontrollable anxiety and fear in patients with GAD.

We also found decreased ReHo in the bilateral SMA, the important knot of the motor system. The SMA is tightly connected with the motor cortex and is thought to be involved in planning of motor action and cognitive control.35,36 One fMRI study37 used psychophysiologic interaction analysis to show that when patients with GAD performed the generalization task, they exhibited enhancement of anterior insula coupling with SMA. Therefore, we suggest that decreased ReHo of the SMA in our study may be related to the impairment of the motor system, muscular tension, and motor restlessness induced by the loss of inhibition reaction in GAD patients.

Our results showed that patients with GAD exhibited increased ReHo in the left STG. Quirk et al38 indicated that the STG was associated with senior cognitive processes with regard to scary experience of humans, as well as the modulation function to activity in the amygdala. Roy et al39 found that the functional connectivities between the amygdala with STG and insula were positively correlated with the anxiety severity scores in the GAD group. In addition, previous studies indicated that the STG in patients with anxiety disorders such as GAD, panic disorder, and social anxiety disorder was atrophied, compared with the control group. The insula and STG play an important role in regulating interoception, which is the sensation of the physiologic condition from the internal organs, such as temperature, pain, and movement, induced by an anxiety response.40 Using voxel-based morphometry, Moon et al41 found reduced volumes of the insula and STG in GAD patients. These findings suggested that the morphologic alterations of the insula and STG were associated with difficulties in emotional response and lower function of interoceptive sense caused by the anxiety symptoms of GAD. Based on the previous findings, we suppose that increased spontaneous neural activity in STG patients with GAD was a compensatory response against anxiety reaction.

We also found that the left MTG showed increased ReHo value. The MTG is involved in many functions, such as language processing, observation of motion, deductive reasoning, and dynamic facial expressions.42–44 Engels et al45 showed activation in the left MTG for anxious apprehension using an emotional Stroop task-based fMRI. Using an automated surface-based approach (FreeSurfer), Strawn et al46 found increased cortical thickness in the left inferior and middle temporal cortex in youth with GAD. The involvement of the middle and inferior temporal gyri in emotional processing was supported by anatomic studies indicating that the posterior temporal cortex received input from visual sensory cortices47 and directed back projections from the amygdala.48 In our study, we found significant positive correlation between the ReHo value in the left MTG and the HAMA score. The more severe the level of anxiety in patients with GAD, the more they demonstrated overt spontaneous activity in the MTG that might involve implicit emotion processing as a compensatory response against anxiety.

Interestingly, we found increased ReHo value in the right SOG. The occipital lobe is the visual information processing center and is responsible for part of language, action feeling, and abstract concepts. A previous study using structural MRI found increased cortical thickness in the occipital region in GAD patients compared to healthy subjects. The finding pointed to the occipital cortex as a region of brain associated with the pathology of and clinical improvement in GAD.49 Other previous studies found that lateral occipital cortical thickness in adolescents with GAD participates in retrograde modulation of the amygdala.50,51 In pediatric patients with GAD, Strawn et al46 also found increased cortical thickness in the right lateral occipital cortex. Furthermore, patients with GAD showed significantly lower activities in the SOG during the use of neutral and anxiety-inducing activities, which can be considered relatively lower ability for visual information processing and deficits of visual attention, compared to healthy controls.52 Considering these previous studies, the increased ReHo in the right SOG in our study might be a compensatory response involving disorders of emotion regulation.

It is noteworthy that there are still several limitations in this study. First, the number of subjects in our study was relatively small, which could affect the statistical analysis and comprehensive interpretation of the findings. A further study including more datasets is needed. Second, although we found abnormalities in the motor cortex, we did not test the motor-related scales. It would be better for future research to use motor-related scales to investigate more fully the decreased spontaneous activity in the motor cortex in patients with GAD. Third, this study is a cross-section preliminary study, and the findings of the study cannot provide longitudinal alteration data on GAD patients.

Conclusion

In conclusion, our results showed altered synchrony of regional spontaneous activity in GAD patients based on rs-fMRI data. The results indicated that altered functional activities are mainly located in the emotional sensorimotor and vision-related regions. In addition, the spontaneous regional activities in the anterior cingulate gyrus, frontal lobe, and temporal lobe may be closely related to the neural mechanism and clinical manifestation of GAD.