Article Text

Download PDFPDF
Research paper
Resting-state fMRI study on drug-naive patients with Parkinson's disease and with depression
  1. ChunYan Luo1,
  2. Qin Chen1,
  3. Wei Song1,
  4. Ke Chen1,
  5. XiaoYan Guo1,
  6. Jing Yang1,
  7. XiaoQi Huang2,
  8. QiYong Gong2,
  9. Hui-Fang Shang1
  1. 1Department of Neurology, West China Hospital, SiChuan University, Chengdu, Sichuan, China
  2. 2Department of Radiology, Huaxi MR Research Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China
  1. Correspondence to Professor Hui-Fang Shang, Department of Neurology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China; hfshang2002{at}126.com

Abstract

Objective This study used resting-state functional MRI (fMRI) to evaluate regional and network alterations in patients with Parkinson's disease (PD) with and without depression.

Method We recruited 29 patients with PD with depression (PD-Dep), 30 patients with PD without depression (PD-NDep), and 30 normal controls. All participants underwent resting-state fMRI scans on a 3-T MR system. The amplitude of low-frequency fluctuation (ALFF) of blood oxygen level-dependent signals was used to characterise regional cerebral function. Functional integration of the brain network was evaluated by seed-based correlation approach.

Results The PD-Dep group showed significantly higher ALFF value in the left orbitofrontal area compared with both the PD-NDep and control groups (p<0.05 corrected by FWE). In patients with PD, the Hamilton Depression Rating Scale score was positively correlated with the ALFF value in the left orbitofrontal cortex (p<0.005 uncorrected). Brain network connectivity analysis revealed reduced functional connectivity of putamen in both PD subgroups. However, the PD-Dep group showed more distributed reduced connectivity in the prefrontal-limbic network than the PD-NDep group did (p<0.05 corrected by FWE).

Conclusions Our study demonstrates that PD-Dep patients are characterised by increased regional spontaneous neural activity in the orbitofrontal area and decreased functional integration within the prefrontal-limbic network. These findings may be helpful for facilitating further understanding of the potential mechanisms underlying depression in PD.

  • Parkinson's Disease
  • Depression
  • Functional Imaging

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Parkinson's disease (PD) is the second most common neurodegenerative disease worldwide, and is characterised by cardinal motor symptoms, including tremor, rigidity, bradykinesia and postural instability.1 However, growing evidence suggests that PD is a disease with numerous non-motor symptoms (NMS) that affect multiple, non-dopaminergic neuronal populations aside from the dopaminergic system.2 ,3 Depression is a prevalent NMS in PD, with a prevalence of approximately 40–50%.4 Although depression may be secondary to progressive and disabling symptoms, several lines of evidence support the concept that depression may be a consequence of the pathologic substrates of the disease.5 Additionally, some studies observed that depression could be the initial symptom in PD other than motor symptoms6; depression in PD is associated with a faster decline of functional and motor abilities, as well as of cognitive performance.7 However, depression, which has been found to be one of the most important factors impairing patients’ quality of life is frequently unrecognised and undertreated.

Substantial efforts have been made in the past decade to elucidate the neural basis of depression in PD. Postmortem studies indicate that pathological processes may already occur in limbic regions during the presymptomatic phase of PD,8 thus providing insight into the pathological mechanisms of depression in PD. Through neurotransmitter imaging, abnormal serotonergic neurotransmission9 ,10 and specific loss of dopamine and noradrenaline innervation in the limbic system11 have been found in PD patients with depression. Abnormal functional activity in the prefrontal area has been characterised as a critical hallmark in previous models of depression pathophysiology.12 In line with these models, altered metabolism or cerebellum blood flow in the frontal cortex has been associated with depression in PD.13–16 With the aid of structural neuroimaging techniques, grey matter loss and white matter abnormalities in the prefrontal, temporal and some limbic regions, were detected in patients with PD with depression. These findings provide a structural basis for these functional changes.17 ,18 Previous studies found that the abnormal brain regions in patients with PD with depression, mainly involved the prefrontal cortex and limbic regions. The outcome highlights a complex pathophysiological mechanism involving deficits in the prefrontal-limbic network in PD patients with depression.

The brain is a network of a large number of different brain regions, each with its own task and function. These regions continuously share information with one other. Depressive manifestation is unlikely to result from a single brain region or neurotransmitter system. Instead, depression can be conceptualised as a multidimensional, system-level disturbance affecting distributed functionally integrated pathways.12 Despite growing evidence of the role of regional deficits in the genesis of depression in PD, little is known about the integrity of the neural network, in which information is continuously processed and transported among structurally and functionally linked brain regions. Thus, examining the human brain as an integrative network of functionally interacting brain regions can provide new insights into the large-scale neural communication in the brain.

A new non-invasive method for assessing regional and neural circuitry function at rest, known as ‘resting-state’ functional MRI (rfMRI), has been recently developed. This method requires minimal patient compliance, avoids potential performance confounders associated with cognitive activation paradigms in task design fMRI research, and is relatively easy to implement in clinical studies. Spontaneous low-frequency (0.01 Hz–0.08 Hz) fluctuations of the blood-oxygen-level-dependent (BOLD) signal observed by rfMRI during the resting state are considered to be physiologically meaningful and related to spontaneous neural activity.19 rfMRI can employ the amplitude of low-frequency fluctuation (ALFF) to detect synchronous regional cerebral activity alterations, and enables the examination of the integrity of brain networks by using functional connectivity (FC) analysis. This technique has been successfully used to detect abnormal functional integration in PD.20–23 However, few experiments have used this technique to explore the abnormal functional integration underlying depression in PD.

Moreover, previous neuroimaging experiments were conducted on patients who have been chronically exposed to anti-Parkinsonism medications. Chronic exposure to anti-Parkinsonism medications may result in the reorganisation of functional integration24 which might not reflect the primary pathophysiological changes induced by PD. Therefore, compared with studies on chronic patients, studies on drug-naive PD patients with depression may be critical to elucidate the underlying mechanism of depression in PD.

This investigation intends to use rfMRI to characterise changes in both regional ALFF intensity and FC of the brain network in drug-naive patients with PD with and without depression, compared with a cohort of normal controls. We investigate the functional changes in these two subgroups of patients to determine whether non-invasive measurements may lead to the identification of the two subgroups at an early stage of clinical intervention. We hypothesised that (1) PD patients with depression show ALFF alterations in the prefrontal or limbic regions, and that (2) different system-level disturbances would be observed in the prefrontal-limbic network between patients with PD with and without depression.

Methods

Participants

The local research ethics committee approved this study, and written informed consent was obtained from all participants. Patients were recruited consecutively from the Movement Disorders Outpatient Clinic of West China Hospital of Sichuan University from January 2010 to November 2012. The recruited patients fulfilled the PD Society Brain Bank diagnostic criteria. From this cohort, patients were excluded if they had (1) moderate to severe head tremor; (2) cerebrovascular disorders, including previous stroke, history of head injury, history of seizure, hydrocephalus, intracranial mass, previous neurological surgery and other neurologic diseases; (3) antidepressant treatment within 6 months prior to the beginning of the study or previous psychiatric therapy; (4) anti-Parkinsonism medications before enrolment; and (5) cognitive impairment. After enrolment, the patients were followed-up for at least 1 year to confirm the diagnosis. Patients with poor response to dopaminergic medication, or who showed emergence of non-Parkinsonism symptoms in the subsequent follow-up visits will be excluded from the study. Diagnosis of depression was established by using the Structured Clinical Interview for DSM-IV Axis I disorders (SCID), administered to each patient by a psychiatrist trained for SCID interview. A total of 29 patients diagnosed as having depressive disorders (PD-Dep group) and 30 patients diagnosed as non-depressed (PD-NDep group) were enrolled. PD-NDep patients were matched with PD-Dep patients on the basis of PD disease severity. Data pertaining to age, gender, handedness, disease duration, and clinical symptom ratings were collected by a movement disorder specialist prior to MRI examination. The severity of PD was evaluated by using the Hoehn & Yahr stage (H&Y) and the Unified PD Rating Scale (UPDRS). Mini-Mental State Examination (MMSE) was used to evaluate cognition. Depression was quantified by using the 24-item Hamilton Depression Rating Scale (HDRS). Functional images were acquired at the same day of clinical assessment. Clinical assessment and image acquisition were conducted prior to the initiation of any treatment.

Additionally, 30 right-handed normal subjects with no history of neurologic or psychiatric diseases were recruited from the friends and spouses of the patients, and were matched for age and gender with the PD subjects.

MRI acquisition

MRI was performed on a 3.0 Tesla MR imaging System (Excite; GE, Milwaukee, Wisconsin, USA) by using an eight-channel phased-array head coil. MR images sensitised to changes in BOLD signal levels (TR=2000 msec, echo time=30 msec, flip angle=90°) were obtained via a gradient-echo echo-planar imaging sequence (EPI). The slice thickness was 5 mm (no slice gap) with a matrix size of 64×64 and a field of view of 240×240 mm2, resulting in a voxel size of 3.75×3.75×5 mm3. Each brain volume comprised 30 axial slices, and each functional run contained 200 image volumes. The fMRI scanning was performed in darkness, and the participants were explicitly instructed to relax, close their eyes and not fall asleep (confirmed by subjects immediately after the experiment) during the resting-state MR acquisition. Earplugs were used to reduce scanner noise, and head motion was minimised by stabilising the head with cushions.

Preprocessing of fMRI data

Functional image preprocessing and statistical analysis were conducted by using SPM (http://www.fil.ion.ucl.ac.uk). The first 10 volumes of functional images were discarded for the signal equilibrium and participant adaptation to scanning noise. The remaining EPI images were preprocessed via the following steps: slice timing, motion correction, spatial normalisation to the standard Montreal Neurological Institute (MNI) EPI template in SPM8, and resample to 3×3×3 mm3, followed by spatial smoothing with an 8 mm full-width at half-maximum (FWHM) Gaussian kernel. According to the record of head motions within each fMRI run, all participants had less than 1.5 mm maximum displacement in the x, y, or z plane, and less than 1.5° of angular rotation about each axis.

ALFF calculation

ALFF maps were calculated by using REST (http://restfmri.net/forum/rest_v17). After preprocessing, the time series for each voxel was filtered (band pass, 0.01 –0.08 Hz) to remove the effects of very low-frequency drift and high-frequency noise, for example, respiratory and heart rhythms. The filtered time series was then transformed to a frequency domain by using fast Fourier transform (FFT) (parameters: taper percent=0, FFT length=shortest). The power spectrum was square root-transformed, and averaged across the frequency of 0.01 Hz–0.08 Hz at each voxel. This averaged square root of activity was taken as the ALFF. For standardisation purposes, the ALFF of each voxel was divided by the global mean ALFF value to standardise data across subjects.

Functional connectivity analysis

Functional connectivity was examined by a seed-based voxel-wise correlation approach. Structural and functional studies have revealed regional deficits in prefrontal limbic-system structure in patients with PD with depression, as well as in patients with general depression. Thus, we selected the following 19 areas as seeds: the left and right dorsal lateral prefrontal areas, orbitofrontal cortex, hippocampus, insula, amygdala, anterior putamen, posterior putamen, caudate and thalamus and the anterior cingulate cortex. We defined the seed areas in the WFU PickAtlas (http://fmri.wfubmc.edu/software) by overlapping the respective template from the automated anatomical labelling (AAL) atlas. Given the uneven amount of dopamine depletion in putamen in PD, which is the most severe in the posterior putamen, we separated the putamen into posterior and anterior parts. The border between these two regions was defined as the line passing through the anterior commissure.21 After bandpass filtering (0.01–0.08 Hz) and linear trend removal, the reference time series for each seed region was extracted by averaging the fMRI time series of all voxels within each region of interest (ROI). Correlation functional analyses were performed by computing the temporal correlation between each seed reference and the rest of the brain in a voxel-wise manner. To remove the possible variances from the time course of each voxel, eight nuisance covariates were regressed, including the white-matter signal, the cerebrospinal fluid (CSF) signal and six head motion parameters. The correlation coefficients in each voxel were then transformed to z-value images by using the Fisher r-to-z transformation to improve normality. Therefore, an entire brain z-value map was created for each subject.

Statistical analysis

Differences between groups in terms of demographic and clinical variables were performed by the Pearson χ2 test, one-way analysis of variance (ANOVA), or the Student t test, as appropriate. A voxel-based comparison of ALFF maps among depressed patients with PD, non-depressed patients with PD, and normal controls was performed by using a design model of one-way ANOVA with age and gender as covariates, followed by posthoc two-sample t tests. The significance threshold was set at p<0.001, and family-wise error (FWE) correction for multiple comparisons was conducted at the cluster level. To assess the relationship between the ALFF value and HDRS score in all patients with PD, the corresponding metrics were entered into the SPM design matrix by using basic models and linear regression analysis. Voxel-wise correlation was assessed at an uncorrected statistical threshold of p<0.005. Only clusters with size greater than 50 were reported. A voxel-based comparison of z-value maps among the three groups was also performed by using a design model of one-way ANOVA with age and gender as covariates, followed by posthoc two-sample t tests. The significance threshold was set at p<0.001. FWE correction for multiple comparisons was also conducted at the cluster level.

Posthoc ROI analysis

The functional network of putamen is presumably affected in both groups of patients. The threshold applied by the whole-brain FC analysis might be too stringent to detect some subtle functional changes. Therefore, we further performed post-hoc ROI analysis to explore whether some revealed changes in FC of putamen are evident in both patient groups. A comparison of the correlation value (z value) among groups was performed by using one-way ANOVA, followed by posthoc two-sample t tests. The statistical significance level was set to p<0.05.

Results

Demographic and clinical characteristics

Demographic and clinical features of the sample were listed in table 1. Age, gender and handedness were not significantly different among the three groups. No significant difference in disease duration, H&Y stage, UPDRS part III score and MMSE score was found between PD-Dep and PD-NDep. By definition, the HDRS score of PD-Dep patients was significantly higher than that of PD-NDep patients (p<0.001).

Table 1

Demographic and clinical characteristics of the total sample

Regional cerebral function

By comparing the whole-brain ALFF maps among three groups, we found a significantly elevated ALFF value in the left orbitofrontal area in the PD-Dep group compared with normal controls (peak coordinates: −27, 36, −6, t value: 5.82; p<0.05 corrected by FWE), as well as in the PD-NDep group (peak coordinates: −27, 36, −6, t value: 5.82; p<0.05 corrected by FWE) (figure 1A,B). No significant decrease in the ALFF value was found in the PD-Dep group. No significant difference was observed between the PD-NDep group and normal controls.

Figure 1

(A) The Parkinson's disease (PD)-Dep group showed higher ALFF value in the left orbitofrontal cortex compared with normal controls. (B) The PD-Dep group showed higher ALFF value in the left orbitofrontal cortex compared with the PD-NDep group (p<0.05, FWE corrected).

In the voxel-wise correlation, we found that the HDRS score of patients with PD was correlated with the ALFF value in the left orbitofrontal cortex (peak coordinates: −45, 42, 0; p<0.005 uncorrected with cluster size >50, figure 2A). The average ALFF value of all voxels within this cluster was determined. Pearson correlation revealed that the mean ALFF was positively correlated with the HDRS score in the whole PD group (r=0.46, p<0.005, figure 2B) and PD-Dep subgroup (r=0.59, p<0.005, figure 2D), but not in the PD-NDep subgroup (r=0.13, p=0.50, figure 2C).

Figure 2

(A) Voxel-wise correlation analysis revealed that the Hamilton Depression Rating Scale (HDRS) score was correlated with the ALFF value in the left orbitofrontal cortex (peak coordinates: −45, 42, 0; p<0.005 uncorrected with cluster size > 50) in patients with Parkinson's disease (PD). (B) Scatterplot shows the significant positive correlation between the HDRS scores and the mean ALFF value in the left orbitofrontal cortex in all patients with PD. (C) Scatterplot shows no significant correlation between the HDRS score and the mean ALFF value in the left orbitofrontal cortex in the PD-NDep group. (D) The scatterplot shows significant positive correlation between the HDRS score and the mean ALFF value in the left orbitofrontal cortex in the PD-Dep group. OFC, orbitofrontal cortex; L, left; R, right.

Neural network function

Functional connectivity analysis was employed to explore alterations in the brain network. Relative to the control group, both the PD-Dep and PD-NDep groups showed significantly reduced connectivity of putamen with mesolimbic regions, particularly in the amygdala, hippocampus, olfactory area and posterior rectus. In addition to reduced connectivity pattern in the mesolimbic-striatal circuit, the PD-Dep group exhibited decreased connectivity among limbic or paralimbic regions, and more distributed decreased connectivity of putamen, for example, temporal gyrus, left precentral gyrus and precuneus. No increased FC was observed in the comparison of PD-Dep or PD-NDep groups with the controls. A direct comparison between the patient groups revealed significantly decreased connectivity between left orbitofrontal cortex and right insular, as well as revealing relative decreased connectivity between left middle temporal gyrus and bilateral putamen in the PD-Dep group than that in the PD-NDep group (table 2 and figure 3A–C).

Table 2

Difference in functional connectivity among PD subgroups and normal subjects

Figure 3

(A) Parkinson's disease (PD)-Dep patients showed decreased connectivity relative to normal controls mainly in the prefrontal-limbic network, temporal-putamen and mesolimbic-putamen circuits. (B) PD-NDep patients showed decreased connectivity relative to normal subjects mainly in mesolimbic-putamen circuits. (C) Significantly decreased connectivity between left orbitofrontal cortex and right insular, as well as relative decreased connectivity between left middle temporal gyrus and bilateral putamen, in the PD-Dep group compared with the PD-NDep group. Blue lines indicate decreased functional connectivity (p<0.001 at voxel level with cluster size >100). AMYG, amygdala; HIP, hippocampus; OLF, olfactory; INS, insular; PUT, putamen; OFC, orbitofrontal cortex; PreCG, precentral gyrus; PCUN, precuneus; Tem, temporal gyrus; LING, lingual gyrus; R, right; L, left.

In the posthoc ROI analysis, we selected the left precentral gyrus, precuneus and temporal gyrus as ROIs. The correlation value (z-value) between posterior putamen and the corresponding ROIs in each MRI scan was extracted. Both patient groups showed reduced FC of putamen with precentral gyrus and precuneus compared with the control group. However, no significant differences in the aforementioned FC alterations between PD-Dep group and PD-NDep group were found. Significantly decreased FC between putamen and temporal lobe was found between the PD-Dep and controls, as well as between the PD-Dep and PD-NDep groups. However, such differences were not observed between the PD-NDep and control groups (figure 4).

Figure 4

Significant reduced functional connectivity in the Parkinson's disease (PD)-Dep group, PD-NDep group or both groups. Significant differences are indicated by asterisks (***, p<0.001; **, p<0.01; *, p<0.05). PP, posterior putamen; PreCG, precentral gyrus; PCUN, precuneus; MTG, mid-temporal gyrus; L, left; R, right; FC, functional connectivity.

Discussion

By using rfMRI, we demonstrated altered regional brain activities and a disruption in the neural network related to mood regulation in drug-naive PD patients with depression. We found a significantly increased ALFF value of the left orbitofrontal cortex in patients with PD with depression than that of patients with PD without depression and normal controls. This finding is coherent with voxel-wise correlation analysis, which revealed a positive correlation between the ALFF value in the left orbitofrontal cortex and the HDRS score in patients with PD. Additionally, we found decreased FC in the prefrontal-limbic circuits and distributed decreased FC of putamen in patients with PD with depression, although decreased FC of putamen was revealed in both groups of patients with PD. By studying a cohort of drug-naive patients with short disease duration, we ruled out potential confounding effects of chronic duration and medication.

The prefrontal cortex is a critical area in PD depression. This area serves an important function as a top-down modulator of emotional tasks.25 Previous studies have shown that lesions in the left prefrontal areas result in depressive symptoms.26 Studies of blood flow and glucose metabolism in patients with PD with depression also reported prefrontal abnormalities,13–16 which mainly involve the orbitofrontal cortex. This finding agrees with the pattern found in primary depression.27 Furthermore, previous studies documented a normalisation of abnormal activity in the prefrontal cortex (more pronounced in the left hemisphere) after treatment with repetitive transcranial magnetic stimulation for depression in PD.28 ,29 The ALFF, measured by the total power within the frequency band of 0.01 Hz–0.08 Hz, provides information on synchronous cerebral activity, and has been widely used in studies of normal and pathological brain functions. By using this measurement, previous studies on depression have documented an increased ALFF value of the inferior frontal cortex in patients with depression. These studies proposed that a higher baseline activity of the frontal cortex might reflect a larger cognitive effort required to exert effective inhibitory control over limbic regions.30 ,31 Our observations of ALFF alterations are in line with these studies and echo previous findings on depressed patients with PD from an alternative perspective. Specifically, our results demonstrate an increased ALFF value in the left orbitofrontal region of patients with PD with depression compared with patients with PD without depression and normal controls, thereby indicating that baseline brain activity impairment of orbitofrontal cortex occurs in patients with PD with depression. Our results highlight the critical role of orbitofrontal region in the depression associated with PD.

By exploring the relationship between depressive symptoms (HDRS score) and ALFF, we found a positive correlation between HDRS score and ALFF value in the orbitofrontal region in patients with PD, reinforcing the hypothesis that this baseline regional activity alteration is associated with depressive symptom in patients with PD. Skidmore et al32 also explored on the relationships between depression and ALFF in patients with PD. However, they found that depression severity positively correlates with ALFF signal in the right subgenual cingulate. Differences in patients’ characteristics (demographics, motor severity, etc) may contribute to the variation. Alternatively, this variation can be explained by the fact that both findings reflect the same pathological changes in the neural circuit associated with depression in patients with PD. The subgenual cortex is strongly connected with the orbitofrontal cortex, the amygdala and the hippocampus, and is suggested as a key conduit of neural traffic between the ‘thinking’ frontal cortex and the phylogenetically older central limbic region.33 Both the subgenual and orbitofrontal cortex are likely to be vital hubs in the neural circuits associated with depression in PD patients.

The limbic system has widespread connections to the prefrontal cortex which has been suggested to regulate the limbic system.34 Current models of mood disturbance are based on the maladaptive functional interactions between limbic and cortical regions that normally are responsible for maintaining homeostatic emotional control in response to cognitive and somatic stress.12 Exploring the changes of FC among those regions in depressed patients with PD might provide us new insight into the mechanism underlying emotional dysregulation of patients with PD. This study provided a first step toward examining the resting-state FC of the prefrontal-limbic network in patients with PD with depression, and revealed decreased FC in this network, particularly between the orbitofrontal cortex and insular. This finding is consistent with that of previous studies on patients with depression that found reduced connectivity in the prefrontal-limbic network.35 Decreased connectivity in the prefrontal-limbic network may result in higher requirement for the frontal cortex to exert effective control over the limbic system, as reflected by the increased ALFF value in the orbitofrontal cortex in our study. Nevertheless, our results confirmed the decreased resting-state connectivity of prefrontal-limbic circuits in a cohort of depressed patients with PD. Our results suggest that such deficient circuits are critical common pathways for the expression of depression with potential relevance to primary mood disorders.

Previous structure imaging studies found that patients with PD with depression had more severer reduction in grey matter density in the anterior cingulate, prefrontal and temporal cortices,17 and reduction in white matter volume in the anterior cingulate and orbitofrontal cortices.18 ,36 Altered metabolism and blood flow in the frontal cortex, anterior cingulate cortex, insular and other limbic regions have been revealed in patients with PD with depression.13–16 ,28 ,29 Additionally, serotonergic, noradrenergic and monoaminergic dysfunction of the limbic system has been documented in PD depression.9–11 Along with our results, these findings highlight that dysfunction of the prefrontal-limbic network is crucial for the development of depressive symptoms in PD.

Aside from the decreased connectivity in prefrontal-limbic circuits, we also observed relatively decreased connectivity between the putamen and middle temporal gyrus in the depressed PD group compared with non-depressed PD group. The middle temporal gyrus is located in the extended dorsal attention system and is involved in working memory, storing new memories and emotion. Structural and functional abnormities in the middle temporal gyrus have been reported in patients with depression.37 ,38 These findings suggest that the middle temporal gyrus is a part of a relevant functional network associated with depression in PD. Involvement of the striatum is also consistently reported in depression.39 ,40 The disconnection between the middle temporal gyrus and putamen in the PD patients with depression revealed by this study might also reflect the disruption of the mood regulation network.

The histopathological hallmarks of PD are the progressive degeneration of the substantial nigral dopamine neurons which results in dopamine depletion in the striatum. The functional network of striatum is presumably affected in patients with PD with and without depression. Indeed, we observed reduced FC of the putamen with other brain regions in all patients, despite finding a more distributed decreased connectivity of putamen in the depressed patients with PD. The connectivity of caudate was relatively intact, which could be explained by the fact that dopamine depletion is less severe in the caudate relative to the putamen. Compared with normal controls, both PD patient groups demonstrated significantly decreased FC of the putamen with the mesolimbic region, which might reflect common pathological conditions in patients with PD. Postmortem studies have indicated that pathological processes may already occur in mesolimbic regions during the presymptomatic phase of PD.8 Functionally deficient mesolimbic-striatal circuits may be associated with the underlying pathology of some early non-motor manifestations in PD. This deficiency might also form the basis of depression in PD and make patients with PD prone to mood disturbance. However, the occurrence of depression might arise from disturbance in more distributed neural networks, for example, functional disruption of the prefrontal-limbic network.

Although decreased FC of putamen with precentral gyrus and precuneus was only evident in the patients with PD with depression in the whole-brain FC analysis, subsequent ROI analysis confirmed a trend of decreased FC of putamen with precentral gyrus and precuneus in patients with PD without depression. This finding indicates that decreased FC of putamen is a common pathological change in all patients with PD. Loss of striatal-sensorimotor cortex coupling has been reported in patients with PD.21 ,22 Precuneus has been suggested to be the ‘core node’ or ‘hub’ of the default mode network (DMN) that is activated during ‘resting consciousness’.23 Functional disruption of the DMN in cognitively unimpaired patients with PD has been reported by a recent study, and is suggested to be correlated with cognitive impairment in PD.23 These network alterations seen in both subgroups of patients with PD may indicate a common pathological condition.

Several issues should be considered when interpreting these results. There is a lack of consensus on the exact physiological nature of ALFF. Although ALFF is thought to reflect spontaneous neural activity, its exact basis is yet to be fully characterised. Nevertheless, ALFF measurement has been demonstrated to reliably distinguish subjects with PD from controls.20 Additionally, information of the method of seed-based correlation analysis used in our study is limited to the functional connections of the selected seed regions, thus making the examination of functional connections patterns on a whole-brain scale difficult. Future studies may benefit from the exploration of the overall organisation of functional communication channels within the brain network, as enabled by the evolving analytic methodology for resting-state data. Finally, the motor disability of the enrolled patients varies. One may consider depression as being reactive to the motor impairment, and postulate that the disruption of the mood network revealed in this study is associated with motor manifestations in PD. However, we strictly selected our patients with matched motor impairment in each group of patients. Therefore, the decreased FC in the prefrontal-limbic network in patients with PD with depression is unlikely to reflect the difference in motor impairment.

In conclusion, our study demonstrates that drug-naive patients with PD with depression are characterised by increased regional spontaneous neural activity in the orbitofrontal area, and decreased FC within the prefrontal-limbic network. Additionally, reduced FC is also more prominent in the middle temporal gyrus-putamen circuit in depressed patients with PD compared with non-depressed patients with PD. While prolonged anti-Parkinsonism medication may result in the reorganisation of functional neural networks through unknown mechanism, our work ruled out the potential confounding effect of chronic duration and medication by studying a cohort of drug-naive patients with short disease duration. This study represents a first step toward examining the functional integrity of prefrontal-limbic-striatal-cortex circuits in PD depression by using resting fMRI. Further studies are required to fully investigate the effect on depressive symptoms in PD.

References

Footnotes

  • CYL and QC have contributed equally to this study.

  • Contributors Acquisition of the data: CYL, QC, WS, KC, XYG, JY; analysis and interpretation of the data: CYL, HF, QC, XQH, QYG; drafting of the manuscript: CYL, QC, HFS; critical revision of the manuscript: QYG, HF.

  • Funding This study was supported by the National Science Fund of China (grant 30973149) and the Science and Technology Bureau Fund of Sichuan Province (grant 2010SZ0069).

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The study was approved by the ethics committee of West China Hospital of Sichuan University.

  • Provenance and peer review Not commissioned; externally peer reviewed.