Study Objective: To assess resting state functional connectivity between the amygdala and cortical sites in Parkinson’s patients who were symptomatic for REM Behavior Disorder.

Design: Between-groups, resting state fcMRI in RBD symptomatic PD patients versus non-symptomatic PD patients

Settings: MAGNETOM Trio 3 Tesla MRI at the Boston VA, and in-person cognitive assessments

Participants: 13 Parkinson’s patients with RBD symptoms and 28 clinically matched PD patients without RBD symptoms were administered neuropsychologic assessments. A subgroup (N=25) of this sample consisting of 11 Parkinson’ patients with RBD symptoms and 14 patients without RBD underwent fcMRI analysis.

Results: The RBD symptomatic group did not significantly differ from the non-RBD group on any background, clinical, or cognitive variables except that a larger percentage of RBD symptomatic patients were on agonists. Seed-analysis of rsf MRI data revealed greater correlations between BOLD time series of the bilateral amygdala and the bilateral precentral, and left transverse temporal regions in the RBD symptomatic group when compared to the RBD non symptomatic group.

Conclusions: RBD dream enactment behaviors, including auditory hallucinations, may be related to BOLD correlations between amygdala and left sided primary auditory cortex.

•    REM behavior disorder
•    Parkinson’s disease
•    Functional connectivity
•    FMRI
•    Amygdala

McNamara P, Minsky A, Clark D, Tripodis Y, Auerbach S, et al. (2016) Resting State Functional Connectivity between the Amygdala and Cortical Sites in Parkinson’s Patients Symptomatic for REM Behavior Disorder. J Sleep Med Disord 3(3): 1051.

PD: Parkinson’s Disease; RBD: REM Behavior Disorder; Rsmri: Resting-State Magnetic Resonance Imaging; LDE: Levadopa Dose Equivalent Dosages; PTSD: Post-Traumatic Stress Disorder; MMSE: Mini Mental State Exam; Moca: Montreal Cognitive Assessment; RBDQ-HK: REM Behavior Disorder Questionnaire Hong Kong; MPRAGE: Magnetization-Prepared Rapid Acquired Gradient Echo; EPI: Echo Planar Image; BOLD: Blood Oxygen Level-Dependent

REM sleep behavior disorder (RBD) is characterized by loss of the atonia normally associated with rapid eye movement or REM sleep [1]. Patients therefore often act out dreams normally associated with REM sleep. They appear to experience visual and auditory hallucinations in association with violent themed dream narratives.

RBD may herald by decades the onset of one of the synucleinopathies such as Parkinson’s Disease (PD), Lewy Body Dementia (LBD) or multiple system atrophy (MSA) [2,3] and appears to be particularly strongly associated with PD [4,5]. Between a third and half of PD patients evidence some degree of RBD symptomology [2]. There is evidence that the presence of RBD in PD patients may indicate a unique disease subtype characterized by greater cognitive impairment and autonomic dysfunction2 . Identification of those PD patients who evidence signs of RBD is therefore an important clinical and research objective.

Functional connectivity analyses of brain regions implicated in RBD and PD have demonstrated utility in identification of biomarkers and predictors of disease states [6]. Among the brain regions implicated in RBD, the amygdala may be key to its clinical phenomenology as it is among the most intensely activated sites during REM sleep [7,8] and the aggressive nature of dreams during REM sleep [9] are consistent with reports of violent or aggressive dreams of patients with RBD. Cortical networks may modulate amygdala reactivity [10,11] but this regulatory influence may be impaired if functional connectivity between the amygdala and cortical sites is degraded. In the current study, we used resting state rsMRI to test the primary hypothesis that functional coordination between amygdala and cortical sites is altered in PD patients with RBD symptoms.

Participants

Thirteen Parkinson’s patients with RBD symptoms and 28 clinically matched PD patients without RBD symptoms were recruited and enrolled into the study (Table 1). The study was approved by the Institutional Review Board at the VA New England Healthcare System-Jamaica Plain campus, and a procedure of informed written consent was followed. PD patients were referred to the study by a Movement Disorders neurologist (R.D.) and the diagnosis was made using the United Kingdom Brain Bank criteria of bradykinesia plus one or more cardinal motor symptoms [12,13]. Patients with Parkinsonism from known causes (e.g., encephalitis, trauma, carbon monoxide exposure, manganese poisoning, hypoparathyroidism, a multi-infarct state, or medications) were excluded. Similarly, other degenerative diseases mimicking PD (e.g., striatonigral degeneration, progressive supranuclear palsy or olivopontocerebellar degeneration) were excluded as well as PD patients with concurrent Alzheimer-like dementia. Most patients evidenced mid-stage disease with mean Hoehn-Yahr [14] score of the entire group 2.37 (Table 1). Thirty-two patients (78.05%) were on some form of levodopa and tested in the ON state. Levodopa dose equivalent dosages (LDE) were calculated for each patient based on previous studies [15] with the mean LDE) for the entire group= 448.23 mg/d (SD = 318.49) which is comparable to other studies in the literature.

Participants were divided into two groups based on presence or absence of RBD symptoms. RBD symptoms were determined by clinical history and self-report using validated questionnaires (described below). Significant RBD symptoms were present by self-report in 13 patients with PD. We were able to examine medical records of the members of this symptomatic group and members of the non-symptomatic group. Ten of the symptomatic group (71%) had a history of depression, but with only one currently on medication (fluoxeteine 20mg/day); 6 (48%) sleep apnea; 5 (35%) post-traumatic stress disorder (PTSD) and 6 (43%) had more than one of these diagnoses. In contrast among the non-symptomatic group 6 (19%) had a history of depression, 6 (19%) sleep apnea, 7 (22%) PTSD and 8 (23%) had more than one of these diagnoses.

All participants were given a background of cognitive and neuropsychological testing including the Mini Mental State Exam (MMSE), the Montreal Cognitive Assessment (MoCA) [16], digit span backwards, matrix reasoning of the WAIS-IV [17] and the STROOP test. Participants were compensated $10/hour for their time testing and $50 if they completed the MRI.

Assessment of RBD risk

Several RBD questionnaires have been developed to quickly and reliably identify RBD symptomology [18]. We used the 13- item RBDQ-HK scale which has demonstrated 82% sensitivity and 87% specificity and a positive predictive value of 86.3% [19]. Scores can range from 0-100, with 19 and above being indicative of RBD. By the RBD-HKQ standards, 13 PD patients in our sample were symptomatic of RBD. These were deemed the at-risk or symptomatic group. To validate the at-risk RBD group, we additionally gave the Parkinson’s Disease Sleep Scale (PDSS) [20] and the Insomnia Severity Index to all participants [21]. We used the insomnia severity index together with medical history and physical examination to identify sleep disturbance as a consequence of a medical or neurological disorder other than PD, mental disorder, medication, or substance abuse. The RBD risk and non-risk groups did not differ on this insomnia index. We compared overall PDSS scores between RBD risk vs non risk groups but also pulled out the three questions relevant to RBD and tabulated them separately between RBD risk and non-risk groups. The PDSS results confirmed the RBD-HKQ results and identified the same at risk subjects. For example, (for the PDSS a lower score indicates greater sleep dysfunction) all of the individuals in the PD RBD-HKQ identified at-risk group scored significantly lower on question 6 regarding distressing dreams than the non-at-risk individuals (RBD risk group mean = 5.79, non-risk mean =8.27; p < .002).

rsfMRI Image Acquisition

Twenty-five participants diagnosed with PD were available to undergo resting state fMRI (rsfMRI). Eleven of those PD participants had been classified as RBD symptomatic and 14 classified as non symptomatic group (NRBD). MRI scans were obtained using a Siemens TIM Trio 3 Tesla MRI with a 12-channel head coil at the VA Boston Healthcare System in Jamaica Plain, Boston, MA, USA. During each scan, two T1-weighted neuroanatomical images were acquired using an MPRAGE (magnetization-prepared rapid acquired gradient echo) pulse sequence [specifications: 176 slices, TR: 2,530 m.s., TE 3.3 m.s. FOV: 256 m.m., flip angle 7°, gaps skip 0.50, with a voxel size of 1*1*1 m.m.] for a total of 6:02 minutes for each full head scan. Additionally, three sets of T2*-weighted functional images of resting state were acquired for each participant using an Echo Planar image (EPI) sequence with a sensitivity to BOLD (blood oxygen level-dependent) contrast (specifications: 38 slices per full head volume, TR: 3,000 m.s., TE: 30 m.s., FOV: 192 m.m., flip angle 90°, 3 m.m. gaps skip 0.8, voxel size 3*3*3.75 m.m.) for a total of 120 volumes and lasting approximately 6:08 minutes for each run. All participants were instructed to keep their eyes open during this time [3].

Image processing was performed in the Free Surfer processing suite [http://surfer.nmr.mgh.harvard.edu/]. Surface models were reconstructed from two T1-weighted scans using automated segmentation of anatomical images. Resting state fMRI scans, per subject, were processed using a standard stream as outlined by Robinson [22] (motion correction, time shifting, concatenation of scans, motion regressed from the time series, regression of the average time course from the white matter and ventricles, band pass filtering between 0.01 and 0.1Hz). Sessions runs, and time points with excessive motion were excluded (30TRs/session; 20TRs/run; 0.5mm/TR; 0.4mm/ motion threshold). The data were smoothed by a surface based procedure with FWHM of 15mm, and each brain was mapped to a Free surfer’sfs average template brain. Seed regions were taken from subcortical segmentations of the bilateral amygdala as outlined by Fischl et al. [23] Vertex-wise partial correlation to the seeds were used for group-level analyses.

Statistical Analysis

All participants were divided into 2 groups: RBD at-risk individuals with an RBD-HKQ score of 19 or greater, and RBD no-risk individuals who scored lower than 19. 13 Parkinson’s patients with RBD symptoms and 28 clinically matched PD patients without RBD symptoms were assessed. We used protected t tests to assess differences on clinical and behavioral measures between the at-risk and non-risk groups.

rsfMRI Statistical Analysis

For imaging analysis, the bilateral amygdala was defined individually per subject in the NRBD (N=14) and the RBD (N=11) group with FreeSurfer’sselxavg3-sess utility. Both within-group and between-group analyses were run with Free Surfer’s mri_ glmfit utility with no covariates, age as a covariate, DRT scores as a covariate, and both age and DRT scores as covariates. Multiple comparisons corrections (MCC) followed Free surfer’s GLM methods utilizing mri_glmfit-sim with a voxel-wise threshold of p < 0.05 and a cluster-wise threshold of p < 0.05 in a 10,000 repetition Monte-Carlo simulation in the final group analysis.

The RBD symptomatic group (composed of 13 symptomatic patients) evidenced a RBD-HKQ score of 32.08 (10.33) while the non-symptomatic group (composed of 28 patients) evidenced an RBD-HKQ of 7.68 (6.18) (t-test=6.34,p<.001). No significant differences emerged on any cognitive tests. Question 6 on the Parkinson’s Disease Sleep Scale, a measure of sleep quality in PD patients, showed a significant difference (RBD symptomatic group mean = 5.79 (2.66); non-symptomatic group mean = 8.27 (2.01); p<.01). Presence of an agonist also showed a significant difference (53.85% of RBD symptomatic group took an agonist, while 10.71% of the non-RBD symptomatic group took an agonist; p<.01). See Table 1.

Table 1: Cognitive Measures for PD Patients.

Within the smaller group that underwent fMRI (N=25) with 14 non-RBD symptomatic and 11 RBD symptomatic PD patients), the mean RBD-HKQ was 8.43 (6.71 sd.) for non-symptomatic group and 31.45 (9.59 sd.) for the symptomatic group; p<.001). No significant differences emerged between the symptomatic and non-symptomatic groups for either of the background clinical, cognitive, or insomnia measures, except for the presence of an agonist (45.45% of RBD-symptomatic subjects- took an agonist, while 7.14% of non-RBD symptomatic subjects took an agonist; p=.026) (see Table 2).
Table 2: Cognitive Measures for PD Patients Who Underwent fMRI.

rsfMRI Results

Simple within-subject bilateral amygdala seed-analysis loaded without covariates revealed widespread regions of positive correlations with the seed in both the NRBD and RBD groups’ rsfMRI acquisitions. Clusters of positive BOLD correlations with the seed were significant (p < 0.005) in NRBD group in the bilateral middle temporal, left precentral, and right medial orbitofrontal, precuneus, and posterior cingulate regions. Clusters of positive BOLD correlations were significant (p < 0.001) in the RBD group in the left fusiform and right superior temporal regions.

Seed-analysis across all 25 PD subjects revealed a direct relationship between increasing RBD scores and increasing positive BOLD correlations with the amygdala seed in the precentral region of the left cortex (p < 0.05). Group analysis between the low scoring NRBD group and the high scoring RBD group found significant differences in BOLD correlations in three regions of the cortex. BOLD correlations with the amygdala seed were significantly greater (p < 0.05) in the RBD group when compared to the NRBD in the bilateral precentral and left transverse temporal regions (see Figure 1).

Figure 1: Regions of significant differences in BOLD correlations to the bilateral amygdala between non-symptomatic and symptomatic RBD participants.

A) The precentral and transverse temporal gyrus of the left hemisphere and the right precentral gyrus showed greater BOLD correlations for RBD symptomatic participants when compared to RBD non-symptomatic participants. B) Linear regression between RBD scores and BOLD correlation coefficients (unit-less) revealed a direct relationship in the precentral cortex of the left hemisphere. C) Distribution of individual participants’ BOLD correlation coefficients for non-symptomatic (N) and non-symptomatic (Y) participants for the precentral and transverse.

No areas in the cortex were found to be significantly anti-correlated to the bilateral amygdala seed in any of these experimental designs.

This is the first study to our knowledge that identifies potential aberrant functional connectivity between the amygdala and cortical sites as a potential contributory factor to symptomology of RBD. BOLD correlations with the amygdala seed were significantly greater (p < 0.05) in the RBD group when compared to the Non-RBD in the bilateral precentral and left transverse temporal regions. The precentral region borders parietal cortex but modulates motor programming. The transverse temporal region encompasses primary auditory cortex encompassing Brodmann’s areas 41 and 42.

Because the cortical sites normally modulate amygdalar functions (among other functional systems) amygdalar reactivity may increase in RBD symptomatic individuals who develop aberrant functional connectivity between the amygdala and the cortical sites.

The amygdala is known to specialize in mediation of threat detection, fear conditioning, and emotional function [24,25]. It maintains reciprocal feedback loops with the anterior insula, temporal lobes including primary auditory cortex, inferior and superior parietal lobes, and the medial, orbital, and dorsolateral prefrontal cortex (DLPFC). Recent neuroimaging studies have demonstrated that cortical hypoactivity in temporal or frontal networks is strongly associated with amygdala hyper activity25and that may be what we see in RBD. With altered functional connectivity between the amygdala and cortical sites, amygdala mediation of aversive emotional information or information concerning perceived threat would no longer be effectively modulated. If that aversive emotional information was normally processed, in part, during dreams associated with REM sleep, then REM sleep dream phenomenology would be accordingly altered in RBD. We speculate that the link between the amygdala and primary auditory cortex documented here may be related to auditory hallucinations experienced during dream enactment behaviors.

RBD is known to precede onset of synucleopathies like PD by several years [4]. We assessed individuals already diagnosed with PD but not RBD and asked if any PD patients exhibited symptoms of RBD. If the patient was symptomatic for RBD we then asked if these symptoms were associated with measureable changes in functional connectivity of brain regions implicated in REM sleep. Our data confirm that RBD symptoms can make their appearance both before onset of PD and after the onset of PD. When RBD symptoms appear in individuals already diagnosed with PD functional connectivity between the amygdala and the transverse temporal region may be compromised. It is therefore important that clinicians working with already diagnosed patients with PD inquire about RBD symptoms as RBD can develop after onset of the disease. We hope to confirm our findings with a larger, multicenter study in the future.

This material is the result of work supported with resources and the use of facilities at the Neuroimaging Research for Veterans Center, VA Boston Healthcare System and was supported by the John Templeton Foundation, grant number 29245.

1. Schenck CH, Bundlie SR, Mahowald MW. REM behavior disorder (RBD): Delayed emergence of Parkinsonism and/or dementia in 65% of older men initially diagnosed with idiopathic RBD, and an analysis of the minimum and maximum tonic and/or phasic electromyographic abnormalities found during REM sleep.Sleep. 2003; 26: A316.

2. Postuma RB, Gagnon JF, Montplaisir JY. REM Sleep Behavior Disorder and Prodromal Neurodegeneration - Where Are We Headed? Tremor Other Hyperkinet Mov (NY). 2013; 3.

3. Postuma RB, Montplaisir J. Potential early markers of Parkinson’s disease in idiopathic rapid-eye-movement sleep behaviour disorder. Lancet Neurol. 2006; 5: 552-553.

4. Boeve BF. REM sleep behavior disorder: Updated review of the core features, the REM sleep behavior disorder-neurodegenerative disease association, evolving concepts, controversies, and future directions. Ann N Y Acad Sci. 2010; 1184:15-54.

5. Boeve BF. Idiopathic REM sleep behaviour disorder in the development of Parkinson’s disease. Lancet Neurol. 2013; 12: 469-482.

6. Hacker CD, Perlmutter JS, Criswell SR, Ances BM, Snyder AZ. Resting state functional connectivity of the striatum in Parkinson’s disease. Brain. 2012; 135: 3699-3711.

7. Dang-Vu TT, Schabus M, Desseilles M, Sterpenich V, Bonjean M, Maquet P. Functional neuroimaging insights into the physiology of human sleep. Sleep. 2010; 33: 1589-1603.

8. Maquet P, Péters J, Aerts J, Delfiore G, Degueldre C, Luxen A, et al. Functional neuroanatomy of human rapid-eye-movement sleep and dreaming. Nature. 1996; 383: 163-166.

9. McNamara P, Johnson P, McLaren D, Harris E, Beauharnais C, Auerbach S. REM and NREM sleep mentation. Int Rev Neurobiol. 2010; 92: 69- 86.

10. Ochsner KN, Gross JJ. The neural architecture of emotion regulation. Handb Emot Regul. 2007; 1:87–109.

11. Banks SJ, Eddy KT, Angstadt M, Nathan PJ, Phan KL. Amygdala-frontal connectivity during emotion regulation. Soc Cogn Affect Neurosci. 2007; 2: 303-312.

12. Gelb DJ, Oliver E, Gilman S. Diagnostic criteria for Parkinson disease. Arch Neurol. 1999; 56: 33-39.

13. Jankovic J. Parkinson’s disease: clinical features and diagnosis. J Neurol Neurosurg Psychiatry. 2008; 79: 368-376.

14. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and mortality. Neurology. 1967; 17: 427-442.

15. Jankovic J. Parkinson’s disease: clinical features and diagnosis. J Neurol Neurosurg Psychiatry. 2008; 79: 368-376.

16. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005; 53: 695-699.

17. Wechsler D. Wechsler adult intelligence scale, fourth ed., San Antonio, TX, NCS Pearson, 2008.

18. Stiasny-Kolster K, Mayer G, Schäfer S, Möller JC, Heinzel-Gutenbrunner M, Oertel WH. The REM sleep behavior disorder screening questionnaire--a new diagnostic instrument. Mov Disord. 2007; 22: 2386-2393.

19. Li SX, Wing YK, Lam SP, Zhang J, Yu MW, Ho CK, et al. Validation of a new REM sleep behavior disorder questionnaire (RBDQ-HK). Sleep Med. 2010; 11: 43-48.

20. Chaudhuri KR, Pal S, DiMarco A, Whately-Smith C, Bridgman K, Mathew R, et al. The Parkinson’s disease sleep scale: a new instrument for assessing sleep and nocturnal disability in Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2002; 73: 629-635.

21. Morin CM, Belleville G, Bélanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep. 2011; 34: 601.

22. Robinson ME, Lindemer ER, Fonda JR, Milberg WP, McGlinchey RE, Salat DH. Close-range blast exposure is associated with altered functional connectivity in veterans independent of concussion symptoms at time of exposure.Human Brain Mapping. 2015; 36: 911- 922.

23. Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. Neuroimage. 1999; 9: 195-207.

24. Roy AK, Shehzad Z, Margulies DS, Kelly AM, Uddin LQ, Gotimer K, et al. Functional connectivity of the human amygdala using resting state fMRI. Neuroimage. 2009; 45: 614-626.

25. Brown VM, LaBar KS, Haswell CC. Altered Resting-State Functional Connectivity of Basolateral and Centromedial Amygdala Complexes in Posttraumatic Stress Disorder. Neuropsychopharmacology. 2014; 39: 361–369.