Functional connectivity and microstructural changes of the brain in primary Sjögren syndrome: the relationship with depression

Abstract

Background

Fatigue and depression are among the most common manifestations of primary Sjögren syndrome (pSS), but information is lacking on the relationship with brain function and microstructural changes.

Purpose

To investigate microstructural changes and brain connectivity in pSS, and to evaluate their relationship with fatigue and depression.

Material and Methods

The study included 29 patients with pSS (mean age 61.2 ± 12.1 years; disease duration 10.5 ± 5.9 years) and 28 controls (mean age 58.4 ± 9.2 years). All the patients completed the Beck’s depression and Fatigue Assessment Scale questionnaires. The imaging protocol consisted of: (i) standard magnetic resonance imaging (MRI) pulse sequences (FLAIR, 3D T1W); (ii) a diffusion tensor imaging pulse sequence; and (iii) a resting state functional MRI pulse sequence. Resting state brain networks and maps of diffusion metrics were calculated and compared between patients and controls.

Results

Compared with the controls, the patients with pSS and depression showed increased axial, radial, and mean diffusivity and decreased fractional anisotropy; those without depression showed decreased axial diffusivity in major white matter tracts (superior longitudinal fasciculus, inferior longitudinal fasciculus, corticospinal tract, anterior thalamic radiation, inferior fronto-occipital fasciculus, cingulum, uncinate fasciculus, and forceps minor-major). Decreased brain activation in the sensorimotor network was observed in the patients with pSS compared with the controls. No correlation was found between fatigue and structural or functional changes of the brain.

Conclusion

pSS is associated with functional connectivity abnormalities of the somatosensory cortex and microstructural abnormalities in major white matter tracts, which are more pronounced in depression.

Keywords

Primary Sjögren syndrome central nervous system magnetic resonance diffusion functional magnetic resonance imaging brain

Introduction

Primary Sjögren syndrome (pSS) is an autoimmune disorder that affects mainly women, at around the age of 55 years (1). The condition is characterized by lymphocytic infiltration and destruction of the exocrine glands, mainly the salivary and the lachrymal glands. Lymphocytic infiltration of visceral organs and vasculitis can give rise to systemic manifestations (1). Involvement of the peripheral nervous system in pSS has been associated with circulating antineuronal antibodies and hypoperfusion due to small-vessel vasculitis (2). Involvement of the central nervous system (CNS) has also been reported at a frequency of 10%–60%, with a range of neurological and psychiatric manifestations, including motor and sensory deficits, fatigue, depression, and cognitive impairment (3). Fatigue and depression are common, and together or independently may have an adverse impact on the quality of life of patients with pSS (4,5). Persistent fatigue is reported in approximately 70% of patients with pSS and depression has been reported in a frequency in the range of 8.3%–75.6% (4). According to a systematic review and meta-analysis, pSS increases the probability of developing depression by 5.36 times (5). Depression has been associated with chronic disease, hypoperfusion, activation of pro-inflammatory cytokines, and increased activity of the immune system; however, in pSS little is known about its etiology (5). The relationship between depression and brain structure and function has never been evaluated in patients with pSS. One study investigated the relationship between fatigue and the volume of gray matter (GM) and white matter (WM) and white matter hyperintensities (WMH) (6). Imaging studies evaluating brain structure in neurologically asymptomatic patients with pSS revealed focal WMH, regional decrease in GM and WM, and extensive loss of WM microstructural integrity with diffusion tensor imaging (DTI) (7 –9). The only resting state functional magnetic resonance imaging (rs-fMRI) study of brain function in pSS showed decreased brain activity in the visual cortex and fronto-parietal junction area (10). Data evaluating both brain structure and function in the same patients with pSS are lacking.

To better investigate the impact of pSS on brain structure and function, we considered it useful to evaluate both in a series of patients with pSS in comparison with control subjects. In addition, because of the high incidence of depression and fatigue in pSS and their adverse impact on the quality of life, we investigated the relationship of brain structure and function with these morbidities in the patients with pSS.

Material and Methods

Study participants

The study group consisted of 29 consecutive, unselected patients with pSS (mean age = 61.2 ± 12.1 years; mean disease duration = 10.5 ± 5.9 years) and 28 matched controls (mean age = 58.4 ± 9.2 years). The diagnosis of pSS was established according to the American-European Consensus Criteria (11). Association with other connective tissue diseases was ruled out and only patients with pSS alone were included in the study. Patients were also excluded if they had a history or clinical signs of cardiovascular disease, peripheral arterial disease, hepatic dysfunction (transaminase levels > 1.5 times the upper limit of normal), renal insufficiency (serum creatinine concentration > 1.6 mg/dL), proteinuria (>0.5 g/d), diabetes mellitus (fasting plasma glucose level ≥126 mg/dL or use of antidiabetic medication), hypertension (arterial blood pressure >140/90 mmHg or use of antihypertensive medication), thyroid-stimulating hormone concentration > 5 mU/mL, dyslipidemia, treatment with corticosteroids during the six months before the study, and a history of neurological problems (autoimmune or demyelinating diseases, migraine, head injury, Alzheimer’s disease). All the patients and controls completed the Beck’s depression scale and the Fatigue Assessment Scale (FAS) questionnaires (12,13). The study complies with the Declaration of Helsinki, was approved by the institutional review board, and all the participants signed a written informed consent agreement.

Imaging protocol

All MRI examinations were performed on the same 1.5-T unit (Gyroscan Intera; Philips Healthcare, Best, The Netherlands) using a quadrature head coil. The imaging protocol in both patients and controls consisted of: (i) single-shot multislice gradient-echo planar imaging, which was used for blood oxygenated level dependent (BOLD) functional images (TR/TE =2000/50 ms, flip angle = 90°, matrix = 64 × 64, slice thickness = 4 mm, gap = 0 mm); each fMRI session consists of 186 scans and lasts 6 min 18 s. During the fMRI scan, the participant was asked to remain awake but still, and not to think about anything in particular; (ii) a single-shot multisection spin-echo echo-planar sequence (TR/TE = 9807/131 ms, field of view [FOV] = 230 mm, matrix = 128 × 128, section thickness = 3 mm, maximum b-value = 700 s/mm², 16 non-collinear diffusion directions, intersection gap = 0 mm) which was used for diffusion tensor imaging (DTI); (iii) T1-weighted high resolution (1 × 1 ×1 mm) three-dimensional (3D) spoiled gradient-echo sequence (TR/TE, 25/4.6 ms, acquisition matrix =256 × 228, FOV = 220 mm) which was used for structural imaging; (iv) axial fluid attenuation inversion recovery (FLAIR) sequence (TR/TE = 6300/120, inversion recovery time = 2150 ms).

Image analysis

White matter lesions

Using the FLAIR images, WM lesions were identified and counted in all study participants. Kruskal–Wallis testing was used to assess differences in the number of lesions between the patients, with or without depression or fatigue, and the controls.

TBSS analysis

Images were processed using the FMRIB Software Library (FSL) software package (FMRIB, Oxford, UK). For each individual, all images (diffusion-weighted and b0 images) were corrected for eddy current-induced distortion and subject motion effect using the FMRIB Diffusion Toolbox (FDT). A brain mask was created from the first b0 image using the Brain Extraction Tool (BET) and FDT was used to fit the tensor model and to compute fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) maps.

Voxel-wise analysis was performed using TBSS. First, the most representative FA image was identified and the FA data from all the individuals were aligned to this target image using the non-linear registration tool FNIRT, which uses a b-spline representation of the registration warp field. Next, the mean FA image was created and thinned to create a mean FA skeleton, which represents the centers of all tracts common to the group. A threshold of FA > 0.2 was applied to the skeleton to include only major fiber bundles. The aligned FA data of each individual were then projected onto this skeleton. Using the non-linear registrations of the FA images, the MD, RD, and AD maps were also projected onto the mean FA skeleton.

Voxel-wise one-way analysis of variance (ANOVA) was applied to assess the FA, MD, RD, and AD differences between the controls and the patients with or without depression or fatigue, with age and gender as covariates. ANOVA analysis with post-hoc between-group comparisons was performed using a permutation-based inference with 5000 permutations. The produced statistical images were thresholded, using the threshold-free cluster-enhancement approach, and corrected for multiple comparisons by controlling the family-wise error rate at the level of P < 0.05. The localization of the significant clusters and their anatomical labeling was based on the Johns Hopkins University WM tractography atlas and the International Consortium for Brain Mapping DTI-81 WM labels.

rs-fMRI analysis

Preprocessing of the resting state data was performed using the Statistical Parametric Mapping (SPM12) software package (14). For each individual, the rs-fMRI volumes were realigned to the first volume, slice-timing corrected, co-registered with structural data, spatially normalized into the standard Montreal Neurological Institute (MNI) space and finally smoothed using a Gaussian kernel of 8 mm FWMH. Group spatial independent component analysis (ICA) was performed using the GIFT toolbox (15). The optimal number of components was chosen according to the minimum descriptive length (MDL) criteria (16). Independent components were calculated using the Infomax algorithm. The back-reconstruction approach (GICA) was used to obtain subject-specific maps and time courses as implemented in GIFT software. Components and their time series were inspected visually and compared with the functional connectivity atlas networks of mialab (http://mialab.mrn.org/data/index.html). Components were discarded if they did not exhibit GM activations and were dominated by high-frequency fluctuations. Individual subject components from the ICA back-reconstruction process were selected to assess between-group effects in SPM12. Specifically, a T-test was used to assess differences between controls and patients. ANOVA analysis with post-hoc between groups comparisons was applied to assess differences between controls and patients with and without depression, and between controls and patients with and without fatigue.

Results

The demographic and clinical characteristics of the patients included in the study are shown in Table 1. None of the prospective participants and controls was excluded for cognitive impairment because they all scored >23 in the mini-mental examination. There was no age difference (T-test, P = 0.292) between pSS patients with (60.4 ± 11.9 years) and without (65.2 ± 10.2 years) depression. Of the 29 patients with pSS, 11 showed depression according to Beck’s scale (>10; mean age = 64.0 ± 14.1 years; mean disease duration = 10.3 ± 6.8 years), all of whom also had a pathological FAS score (>21). Of the 18 patients with pSS without depression (mean age = 59.5 ± 10.8 years; mean disease duration = 10.6 ± 5.6 years), five showed a pathological FAS score. None of the controls presented depression according to Beck’s scale or fatigue according to the FAS score.

Table 1.

Demographic and clinical characteristics of 29 study patients with primary Sjӧgren syndrome.

Age (years)	61.6 ± 12.0
Sex (F/M)	29/0
Disease duration (years)	10.9 ± 6.2
Positive salivary gland biopsy (%)	100
Arthritis/arthralgia (%)	68.9
Raynaud phenomenon (%)	58.6
Depression (%)	37.9
Fatigue (%)	55.2

Values are given as mean ± SD unless otherwise specified.

The number of WMH was larger in the patients with pSS (median = 10; range = 1–120) than in the controls (median = 4; range = 1–39) (P = 0.00022). Positive correlation was demonstrated between the number of WMH and MD (0.05 > P > 0.016) and RD (P = 0.05) in patients with pSS in the anterior thalamic radiation, the corticospinal tract, the cingulum, the forceps minor and major, the inferior fronto-occipital fasciculus, the superior and inferior longitudinal fasciculus, and the uncinate fasciculus.

Compared with the controls, the patients with pSS and depression presented higher MD, RD, and AD and lower FA (Figs. 1 and 2, Table 2) while the patients without depression showed decreased AD in the same major WM tracts (Table 3). The patients with pSS with depression presented higher MD, RD, and AD and lower FA values in the same major WM tracts than those without depression (Table 4). No differences in the DTI metrics were found between the controls and patients with or without fatigue.

Fig. 1.

Colored statistical maps thresholded at P < 0.05 (corrected) and overlaid on the Montreal Neurological Institute brain template, showing areas with increased radial diffusivity (RD) in patients with primary Sjӧgren syndrome with depression, compared with controls.

Fig. 2.

Colored statistical maps thresholded at P < 0.05 (corrected) and overlaid on the Montreal Neurological Institute brain template, showing areas with increased axial diffusivity (AD) in patients with primary Sjӧgren syndrome with depression, compared with controls.

Table 2.

Location and size of brain areas (clusters) with the most significant increase in AD, RD, MD, and decrease in FA in patients with primary Sjӧgren syndrome with depression, compared with controls, using tract-based spatial statistics.

	Size (voxels)	P(max)	X(max) (mm)	Y(max) (mm)	Z(max) (mm)	Anatomic region
AD	3506	0.008	26	20	7	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus, hippocampus), forceps (major, minor), inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
	3325	0.016	–24	–11	35	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus), forceps minor, inferior fronto-occipital fasciculus, superior longitudinal fasciculus, uncinate fasciculus
	1095	0.015	–28	–49	24	Anterior thalamic radiation, corticospinal tract, cingulum (hippocampus), forceps major, inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
RD	47999	0.001	18	–44	40	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus, hippocampus), forceps (major, minor), inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
MD	39337	0.001	17	10	–16
FA	40896	0.005	18	–42	25

The probability P(max) and the coordinates X(max), Y(max), and Z(max) in the MNI standard space of the most significant voxel are also displayed.

AD, axial diffusivity; FA, fractional anisotropy; MD, mean diffusivity; MNI, Montreal Neurological Institute; RD, radial diffusivity.

Table 3.

Location and size of brain areas (clusters) with the most significant decrease in AD, in patients with primary Sjӧgren syndrome without depression, compared with controls, using tract-based spatial statistics.

Size (voxels)	P(max)	X(max) (mm)	Y(max) (mm)	Z(max) (mm)	Anatomic region
AD
5818	0.019	–35	–46	27	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus), Forceps (major and minor), inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
1727	0.038	27	22	26	Anterior thalamic radiation, cingulum (cingulate gyrus), forceps minor, inferior fronto-occipital fasciculus, superior longitudinal fasciculus, uncinate fasciculus

The probability P(max) and the coordinates X(max), Y(max), and Z(max) in the MNI standard space of the most significant voxel are also displayed.

AD, axial diffusivity; MNI, Montreal Neurological Institute.

Table 4.

Location and size of brain areas (clusters) with the most significant increase in AD, RD, MD, and decrease in FA in patients with primary Sjӧgren syndrome with depression (n = 11), compared with patients with pSS without depression (n = 18), using tract-based spatial statistics.

	Size (voxels)	P(max)	X(max) (mm)	Y(max) (mm)	Z(max) (mm)	Anatomic region
AD	38996	0.001	15	12	–19	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus, hippocampus), forceps (major, minor), inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
RD	51248	0.001	28	17	–2
MD	51821	<0.001	26	17	–8
FA	2451	0.028	–32	–66	2	Anterior thalamic radiation, cingulum (hippocampus), forceps (major, minor), inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
	1599	0.041	–12	30	10	Anterior thalamic radiation, cingulum (cingulate gyrus), forceps minor, inferior fronto-occipital fasciculus, inferior and superior longitudinal fasciculus, uncinate fasciculus
	1275	0.041	25	27	3	Anterior thalamic radiation, corticospinal tract, cingulum (cingulate gyrus), forceps minor, inferior fronto-occipital fasciculus, superior longitudinal fasciculus, uncinate fasciculus

The probability P(max) and the coordinates X(max), Y(max), and Z(max) in the MNI standard space of the most significant voxel are also displayed.

AD, axial diffusivity; FA, fractional anisotropy; MD, mean diffusivity; MNI, Montreal Neurological Institute; RD, radial diffusivity.

The patients showed decreased brain activation compared to the controls in the sensorimotor network (cluster size = 52,433 mm³; MNI coordinates of the most significant voxel = [6.3, 4.6, 56.7], P < 0.001; Fig. 3). No differences in functional connectivity were found between the patients with or without depression or fatigue and the controls.

Fig. 3.

Resting state functional magnetic resonance imaging colored maps overlaid on the Montreal Neurological Institute brain template showing decreased brain activation in the sensorimotor network in patients with primary Sjӧgren syndrome, compared with controls.

Discussion

The major findings of the present study on pSS were the association of depression with microstructural changes in the brain and the decreased activation of the sensory motor network.

Macro- and microstructural changes

Several studies of patients with pSS have suggested an underlying vascular disease leading to hypoperfusion as one of the major causes of CNS involvement (8,17–19). WMH have been associated with brain vascular disease and, in agreement with previous reports on pSS, in this study a higher number of WMH was observed in the patients with pSS than in the controls (8,18,20). In pSS, neuropathological studies have demonstrated small-vessel vasculopathy and infarction (17,18), while single photon emission computed tomography (SPECT) and positron emission tomography (PET) studies have demonstrated hypo-perfusion and decreased glucose metabolism respectively (21,22).

Microstructural changes of the WM have been reported in patients with pSS (9), but this is the first study to evaluate DTI metrics in relation with depression and fatigue in patients with pSS. Lower AD was seen in the patients without depression and increased MD, RD, and AD and decreased FA were demonstrated in patients with depression. Several pathogenetic mechanisms may account for these findings. First, Wallerian degeneration of the WM tracts related to cortical and subcortical GM atrophy. Atrophy of the frontal, parietal, occipital, and cerebellar cortices, and of the thalami and caudate nuclei has previously been demonstrated in pSS (8). In the initial stages of Wallerian degeneration, axonal disintegration creating multiple barriers along the long axis of the WM tracts is predominant, and appears with decreased AD, while a less prominent myelin fragmentation does not always decrease the RD (23,24). In the later stages of Wallerian degeneration, axonal atrophy, demyelination, and increased water content appear and these explain the increase in AD, RD, and MD and the decrease in FA. A second mechanism is hypoperfusion in pSS, which has been attributed mainly to small-vessel vasculitis (19). Another cause of brain hypoperfusion in pSS could be dysregulation of the cytokine network, reflected by local and systemic overexpression of pro-inflammatory cytokines which affect the vascular endothelium, causing alteration in the vasomotor tone (25 –27). Hypoperfusion of WM may lead to apoptotic death of oligodendrocytes, which in turn results in hypomyelination (23). Hypomyelination and multiple small infarcts may increase primarily RD and secondarily MD. Third, direct damage of the WM via autoantibodies against myelin (anti-aquaporin-4 antibodies) may also contribute to demyelination and axonal loss, and thus affect the AD, RD, MD, and FA (9). Finally, all these mechanisms may coexist and contribute in varying degrees to the microstructural changes of the WM. In the present study the relationship between WMH and DTI metrics was further investigated, and a positive association was found with MD and RD. The stronger correlation with RD is probably related to hypomyelination and increased water content.

A lack of correlation was observed between fatigue and the number of WMH or the DTI metrics. This is in agreement with a previous study of pSS which reported no relationship between fatigue and GM and WM volume and WMH (6). A possible explanation would be that fatigue is part of the systematic non-CNS-related manifestations of pSS.

No correlation was found between depression and the number of WMH, but the patients with depression presented more pronounced changes in the DTI metrics of major WM tracts serving cortical areas related to attention, memory, and emotion. Specifically, the superior longitudinal fasciculus facilitates the formation of a bidirectional neural network that is necessary for core processes such as attention, memory, and emotion (28). Abnormalities of the inferior longitudinal fasciculus and the inferior fronto-occipital fasciculus have been associated with thought disorders, visual emotion, and cognitive impairment (29,30). The uncinate fasciculus is considered to belong to the limbic system, which is a critical structure in emotion and memory (30). The anterior thalamic radiation has been associated with mood regulation and equilibrium between positive and negative affective states (31,32). In accordance with the above findings, disrupted global integrity and regional connectivity of multiple brain networks involved in attention, memory, processing and regulation of emotion, and decreased FA in the same major WM tracts have been reported in patients with depression (30,33,34). Furthermore, human and animal studies have suggested that both pro-inflammatory cytokines and alteration of oligodendrocytes play an important role in the pathophysiology of depression (27). The same mechanisms have also been implicated in the CNS pathology of pSS suggesting that the two conditions might be interrelated (25). Because of the similarity of the affected WM tracts between pSS and depression, it is difficult to distinguish whether the observed microstructural changes in the present study are secondary to the neuropathology of pSS or they represent prodromal findings of depression. In the last case, depression would represent a co-morbidity necessitating additional appropriate therapeutic regimens.

Functional changes

A study of patients with pSS evaluating the resting state activity of the brain using ReHo analysis showed decreased coactivation in the sensorimotor cortex compared with that of controls, suggesting weakened synchronization and decreased resting state activity in pSS (10). In agreement with this, the present study, using ICA analysis of rs-fMRI data, also revealed decreased brain activation in the sensorimotor network. The advantages of ICA are that it provides information on both short- and long-range connections across brain regions and does not overestimate brain activity (35). The pre- and post-central gyrus constitute the motor and sensory areas of the brain and represent the origin of the corticospinal tract. The present study detected both decreased brain activation in the sensorimotor network and abnormal WM microstructure in the corticospinal tract. These functional and microstructural abnormalities could be explained by several mechanisms: (i) retrograde Wallerian degeneration of the corticospinal tracts due to peripheral neuropathy could be the underlying mechanism, leading to decreased activation of the somatosensory cortex (2); (ii) primary degeneration of the somatosensory cortex, reported as decreased volume in a previous study on pSS (8); (iii) WM involvement appearing in the present study as WMH and microstructural changes may affect connectivity between cortical and subcortical GM; and (iv) a combination of all these mechanisms.

The main limitations of the present study are the small number of patients with depression or fatigue and the 1.5-T field strength of the MRI. Future studies, with a larger number of participants and higher magnetic fields, would be useful to validate the potential of functional and structural MRI for improving our understanding of the mechanisms by which pSS affects the brain.

In conclusion, patients with pSS may present decreased functional connectivity of the somatosensory cortex and microstructural changes in the corticospinal tracts and major WM tracts, which seem more pronounced in patients with depression. Further studies are necessary to evaluate whether depression is secondary to the neuropathology of pSS or represents a co-morbidity.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Loukas G Astrakas

Maria I Argyropoulou

References

Haugen

Peen

Hulten

, et al. Estimation of the prevalence of primary Sjogren’s syndrome in two age-different community-based populations using two sets of classification criteria: the Hordaland Health Study. Scand J Rheumatol 2008; 37:30–34.

Pavlakis

Alexopoulos

Kosmidis

, et al. Peripheral neuropathies in Sjogren syndrome: a new reappraisal. J Neurol Neurosurg Psychiatry 2011; 82:798–802.

Margaretten

Neurologic manifestations of primary Sjogren syndrome. Rheum Dis Clin

North Am 2017; 43:519–529.

Bowman

SJ.

Primary Sjogren’s syndrome: too dry and too tired. Rheumatology (Oxford) 2010; 49:844–853.

Cui

Yin

, et al. Depression in primary Sjogren’s syndrome: a systematic review and meta-analysis. Psychol Health Med 2018; 23:198–209.

Hammonds

Lauvsnes

Dalen

, et al. No structural cerebral MRI changes related to fatigue in patients with primary Sjogren’s syndrome. Rheumatol Adv Pract 2017; 1:rkx007.

Segal

Mueller

Zhu

, et al. Disruption of brain white matter microstructure in primary Sjogren’s syndrome: evidence from diffusion tensor imaging. Rheumatology (Oxford) 2010; 49:1530–1539.

Tzarouchi LC, Tsifetaki N, Konitsiotis S, et al. CNS involvement in primary Sjogren syndrome: assessment of gray and white matter changes with MRI and voxel-based morphometry. AJR Am J Roentgenol 2011; 197:1207–1212.

Tzarouchi LC, Zikou AK, Tsifetaki N, et al. White matter water diffusion changes in primary Sjogren syndrome. AJNR Am J Neuroradiol 2014; 35:680–685.

10.

Xing

Shi

Leng

, et al. Resting-state fMRI in primary Sjogren syndrome. Acta Radiol 2018; 59:1091–1096.

11.

Goules

Tzioufas

Moutsopoulos

HM.

Classification criteria of Sjogren’s syndrome. J Autoimmun 2014; 48–49:42–45.

12.

Wang

Gorenstein

Assessment of depression in medical patients: a systematic review of the utility of the Beck Depression Inventory-II. Clinics (Sao Paulo) 2013; 68:1274–1287.

13.

Zyga

Alikari

Sachlas

, et al. Assessment of fatigue in end stage renal disease patients undergoing hemodialysis: prevalence and associated factors. Med Arch 2015; 69:376–380.

14.

Penny

Friston

Ashburner

, et al., editors. Statistical Parametric Mapping. The Analysis of Functional Brain Images. Cambridge, MA: Academic Press, 2007.

15.

Calhoun

Adali

Pearlson

, et al. A method for making group inferences from functional MRI data using independent component analysis. Hum Brain Mapp 2001; 14:140–151.

16.

Adali

Calhoun

VD.

Estimating the number of independent components for functional magnetic resonance imaging data. Hum Brain Mapp 2007; 28:1251–1266.

17.

Alexander

EL.

Neurologic disease in Sjogren’s syndrome: mononuclear inflammatory vasculopathy affecting central/peripheral nervous system and muscle. A clinical review and update of immunopathogenesis. Rheum Dis Clin North Am 1993; 19:869–908.

18.

Alexander

Beall

Gordon

, et al. Magnetic resonance imaging of cerebral lesions in patients with the Sjogren syndrome. Ann Intern Med 1988; 108:815–823.

19.

Alexander

Ranzenbach

Kumar

, et al. Anti-Ro(SS-A) autoantibodies in central nervous system disease associated with Sjogren’s syndrome (CNS-SS): clinical, neuroimaging, and angiographic correlates. Neurology 1994; 44:899–908.

20.

Morgen

McFarland

Pillemer

SR.

Central nervous system disease in primary Sjogrens syndrome: the role of magnetic resonance imaging. Semin Arthritis Rheum 2004; 34:623–630.

21.

Huang

Chiu

Kao

, et al. Detecting abnormal regional cerebral blood flow in patients with primary Sjogren’s syndrome by technetium-99m ethyl cysteinate dimer single photon emission computed tomography of the brain–a preliminary report. Rheumatol Int 2003; 23:174–177.

22.

Kao

Lan

, et al. Regional cerebral blood flow and glucose metabolism in Sjogren’s syndrome. J Nucl Med 1998; 39:1354–1356.

23.

Winklewski

Sabisz

Naumczyk

, et al. Understanding the physiopathology behind axial and radial diffusivity changes-what do we know? Front Neurol 2018; 9:92.

24.

Pierpaoli

Barnett

Pajevic

, et al. Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. NeuroImage 2001; 13:1174–1185.

25.

Roescher

Tak

Illei

GG.

Cytokines in Sjogren’s syndrome: potential therapeutic targets. Ann Rheum Dis 2010; 69:945–948.

26.

Figueroa-Vega

Moreno-Frias

Malacara

JM.

Alterations in adhesion molecules, pro-inflammatory cytokines and cell-derived microparticles contribute to intima-media thickness and symptoms in postmenopausal women. PLoS One 2015; 10:e0120990.

27.

Yang

Gao

Zhang

, et al. Changes in proinflammatory cytokines and white matter in chronically stressed rats. Neuropsychiatr Dis Treat 2015; 11:597–607.

28.

Kamali

Flanders

Brody

, et al. Tracing superior longitudinal fasciculus connectivity in the human brain using high resolution diffusion tensor tractography. Brain Struct Funct 2014; 219:269–281.

29.

Ashtari

Anatomy and functional role of the inferior longitudinal fasciculus: a search that has just begun. Dev Med Child Neurol 2012; 54:6–7.

30.

Lai

YT.

The white matter microintegrity alterations of neocortical and limbic association fibers in major depressive disorder and panic disorder: the comparison. Medicine (Baltimore) 2016; 95:e2982.

31.

Coenen

Panksepp

Hurwitz

, et al. Human medial forebrain bundle (MFB) and anterior thalamic radiation (ATR): imaging of two major subcortical pathways and the dynamic balance of opposite affects in understanding depression. J Neuropsychiatry Clin Neurosci 2012; 24:223–236.

32.

Lai

YT.

Alterations in white matter micro-integrity of the superior longitudinal fasciculus and anterior thalamic radiation of young adult patients with depression. Psychol Med 2014; 44:2825–2832.

33.

Gong

Hou

Wang

, et al. Disrupted topology of hippocampal connectivity is associated with short-term antidepressant response in major depressive disorder. J Affect Disord 2018; 225:539–544.

34.

Pasternak

Kelly

Sydnor

, et al. Advances in microstructural diffusion neuroimaging for psychiatric disorders. NeuroImage 2018; 182:259–282.

35.

Rosazza

Minati

Resting-state brain networks: literature review and clinical applications. Neurol Sci 2011; 32:773–785.