Neurochemical Features of Rem Sleep Behaviour Disorder

Dopaminergic deficiency, shown by many studies using functional neuroimaging with Single Photon Emission Computerized Tomography (SPECT) and Positron Emission Tomography (PET), is the most consistent neurochemical feature of rapid eye movement (REM) sleep behaviour disorder (RBD) and, together with transcranial ultrasonography, and determination of alpha-synuclein in certain tissues, should be considered as a reliable marker for the phenoconversion of idiopathic RBD (iRBD) to a synucleopathy (Parkinson’s disease –PD- or Lewy body dementia -LBD). The possible role in the pathogenesis of RBD of other neurotransmitters such as noradrenaline, acetylcholine, and excitatory and inhibitory neurotransmitters; hormones such as melatonin, and proinflammatory factors have also been suggested by recent reports. In general, brain perfusion and brain glucose metabolism studies have shown patterns resembling partially those of PD and LBD. Finally, the results of structural and functional MRI suggest the presence of structural changes in deep gray matter nuclei, cortical gray matter atrophy, and alterations in the functional connectivity within the basal ganglia, the cortico-striatal, and the cortico-cortical networks, but they should be considered as preliminary.


Introduction
In 1986, Schenk et al. [1,2] described a new type of parasomnia in patients (most of them men) with several neurological diseases, resembling previous findings in studies in cats with pontine tegmental lesions, that consisted in abnormal behaviors during rapid eye movement (REM) sleep such as "stereotypical hand motions, reaching and searching gestures, punches, kicks, and verified dream movements". Polysomnography (PSG) showed variable loss of chin atonia, increased REM limb-twitch activity, increased REM ocular activity, and density and increased duration of stage 3-4 slow-wave sleep. Clonazepam and desipramine improved considerably these symptoms. This disorder was designated as "REM sleep behaviour disorder" (RBD). A review of diagnostic tools for RBD including 58 studies concluded that sleep history might be sufficient for the diagnosis of RBD, although PSG study should be necessary for a definitive diagnosis, while the value of objective measurements including visual electromyographic (EMG) scoring methods, actigraphy, cardiac 123 I-metaiodobenzylguanidine ( 123 I-MIBG) scintigraphy, and behavioral classification and video analysis was not established [3].
Olson et al. [4] described a large series (93 patients, 87% men) of patients with RBD and found the presence of neurological disorders in 57% of them (all but 14% of them with Parkinson's disease -PD-dementia without parkinsonism, and multiple system atrophy -MSA), and described that RBD preceded parkinsonism in 52% of PD patients (although the first description of RBD preceding PD was done by Tan & Salgado [5]).
In general, conventional neuroimaging showed non-specific alterations, and in 87% of patients, RBD improved totally or partially with clonazepam. Other group described that 45% of 44 patients diagnosed with RBD developed a neurological disorder (mainly PD and Lewy Body dementia-LBD) after a mean of 11.5 years from the reported onset of RBD and a mean follow-up of 5.1 years from the diagnosis of idiopathic RBD (iRBD) [6], and 82% of them (16 with PD, 14 with LBD, 1 with MSA, and 5 with mild cognitive impairment) at 14 years of follow-up [7].
On the other hand, Boeve et al. [8], in a series of 398 patients with parkinsonism and/or cognitive impairment, described a significantly higher frequency of probable RBD and PSG confirmed RBD in patients with synucleopathies (PD, LBD, or MSA) compared with those without synucleopathies (progressive supranuclear palsy -PSP-, corticobasal degeneration -CBD-, frontotemporal dementia -FTD-, Alzheimer's disease -AD-, mild cognitive impairment -MCI-, primary progressive aphasia -PPA-or posterior cortical atrophy -PCA).
From the etiological point of view, RBD is usually classified as "idiopathic" (iRBD, when at the time of diagnosis there is no evidence of diagnosed neurological disease) or secondary. Secondary RBD is usually related to a previously diagnosed neurodegenerative disease, narcolepsy, autoimmune disease, or induced by drugs [9,10], but also related to structural lesions affecting the pons, medulla, or limbic system [11][12][13]. However, as it should be discussed later, many of the patients diagnosed with iRBD, after a long-term follow up, develop neurodegenerative diseases, mainly PD, LBD, or MSA.
Regarding the etiology of iRBD, only a few studies addressed the possible role of genetic and environmental factors: (1) A first study identified a hexanucleotide repeat expansion in the C9orf72-SMCR8 complex subunit (C9orf72) gene (chromosome 9p21.2; gene ID 203228, MIM 614260; this gene encodes a protein with an important role in the regulation of endosomal trafficking), which has been related with familial amyotrophic lateral sclerosis and FTD dementia, in 2 of 344 patients diagnosed with RBD [15]. (2) Missense variations in the glucosylceramidase beta (glucocerebrosidase or GBA) gene (chromosome 1q22; gene ID 2629, MIM 606463), related with PD and DLBD, have been found in 7 of 69 iRBD patients (11.6%) and in 1 of 84 healthy matched controls (1.2%, p = 0.026) [16].  [17]. (4) A study involving 347 RBD patients and 347 matched controls showed that, compared with controls, RBD patients were more likely to smoke, to report a previous head injury, and to have worked as farmers, with a borderline increase in welding, to have previously occupational exposure to pesticides, and to have few years of formal schooling, while there were no significant differences in coffee consumption [18].
This narrative review focuses in provide an extensive description of studies published related with the neurochemistry and biochemical findings of RBD.

Search Strategy
The references used for this review were identified through a PubMed search which including the period from 1966 until 31 July 2021. The term "REM sleep behavior disor-der" was crossed with "neurochemistry" (5 items), "biochemistry" (22 items), "neurotransmitters"(321 items), "dopamine" (326 items), "noradrenaline" (32 items), "norepinephrine" (32 items), "serotonin" (55 items), "acetylcholine" (22 items), "GABA"(54 items), "gammaaminobutyric acid"(18 items), "histamine" (2 items), "glycine"(15 items), "glutamate" (17 items), "neuropeptides"(39 items), "white matter"(14 items), "gray matter"(23 items), "cortical thickness"(10 items), "neuroimaging" (176 items), and "magnetic resonance neuroimaging" (66 items). We made a selection (after examining one-for-one) of the references strictly related to the neurochemical findings in RBD (a total of 180) from the 537 references retrieved by the whole search. Table 1 summarizes the results of functional neuroimaging studies with Single Photon Emission Computerized Tomography (SPECT) or Positron Emission Tomography (PET) reported to date in patients diagnosed with RBD, including those using different methods and tracers for the presynaptic dopaminergic (DA) terminal and postsynaptic dopamine D 2 receptors (DRD 2 ). The majority of studies analyzing the presynaptic DA terminal have shown a significant decrease in the striatal tracer uptake in many patients diagnosed, at least initially, with iRBD , with some exceptions [45], being this decrease lower than that found in patients diagnosed with PD [27,28,31,43,45] or LBD [43] with or without concomitant RBD, although PD patients with versus those without RBD did not show significant differences [46]. However, patients with and without decreased striatal tracer uptake did not differ in clinical features according to one study [32], while in others, iRBD patients with mild motor impairment or with higher Unified Parkinson's Disease Rating Scale (UPDRS) motor score showed decreased striatal tracer uptake when compared with those without mild motor impairment and with controls [33,34]. Patients with iRBD and MCI showed more frequently decreased striatal tracer uptake [37]. Follow-up studies have shown that many patients initially diagnosed with iRBD and reduced striatal tracer uptake at the presynaptic DA terminal at baseline developed neurodegenerative disorders such as PD, DLB, and MSA [21,22,30,40]. In contrast, studies on postsynaptic DRD 2 have not shown significant differences in the tracer uptake in patients with iRBD compared with controls [38,41]. A meta-analysis of studies published up to 2018 on the presynaptic dopaminergic terminal neuroimaging in RBD patients showed that tracer uptake progressively decreased from controls to iRBD and eventually PD patients with RBD at the putamen level, while tracer uptake at caudate overlapped between patients with iRBD and those with PD without RBD [47].

Dopaminergic Dysfunction
Substantia nigra (SN) hyperechogenicity (SNH), assessed by transcranial sonography (TCS), a finding that likely reflects increased iron content in the SN, is considered as a useful diagnostic marker for nigrostriatal degeneration in PD. Because an important percentage of patients initially diagnosed with iRBD develop PD or other synucleopathies, this technique should be a useful tool to predict conversion of iRBD to PD and other neurodegenerative diseases. SNH was significantly more frequent in patients diagnosed with iRBD than in controls [21,[48][49][50], although less frequent than in patients diagnosed with PD [48,51], and had a similar frequency than that reported in MSA patients [51]. SNH was more frequent in iRBD patients with than in those without mild motor abnormalities [24]. Iranzo et al. [21] found SNH in 36% of 39 patients with iRBD and in 11% of 149 controls, and described a sensitivity of 100% and a specificity of 55% of the combined use of TCS with 18 F-N-(3fluoropropyl)-2beta-carbon ethoxy-3beta-(4-iodophenyl) nortropane ( 123 I-FP-CIT) SPECT for the prediction of conversion of iRBD to synucleopathies after 2-5 years of followup, while the use of TCS alone had a sensitivity of 42.1% and a specificity of 67.7% to make this prediction after 5 years of follow-up [52]. Miyamoto et al. [53] described a higher phenoconversion rate to other synucleopathies in patients with SNH (57.4%) than in those with normal TCS (25.0%). Patients with SNH showed decreased 6-18 F-fluorometatyrosine (FMT) striatal uptake compared with those with normal TCS in another study [44]. In addition, an important percentage of iRBD patients (50%) showed basal ganglia hyperechogenicity, a finding that was less frequent in PD patients (18.2%) and more frequent in MSA patients (66.7%) [51]. SNH, combined with hypoechogenicity of the brainstem raphe, has been described as useful to detect comorbid depression in iRBD patients [54].
It has been described an anecdotal report of RBD induced by quetiapine [55]. This drug acts as DRD 1 and DRD 2 antagonist, but also as serotonin (5-hydroxytryptamine-5-HT) 1 and 2 (5-HT 1 and 5-HT 2 ), histamine H 1 , and α1 and α2-adrenergic receptors antagonist. Therefore, it is unknown if the ability of this drug to induce RBD could be related to the dopaminergic system. Rats with hemiparkinsonism induced by unilateral lesions of the nigrostriatal system with 6-hydroxydopamine (6-OH-DA) showed when compared with normal rats, more REM epochs with muscle activity similar to that found in PD patients with sleep disorders including RBD, suggesting a role of dopaminergic deficit in the pathogenesis of RBD in this model [56]. This is also suggested by the description of improvement of RBD symptoms in PD patients under therapy with levodopa [5,57] or with the monoamine oxidase B (MAOB) inhibitor selegiline [58], and the improvement of iRBD with the dopamine agonist pramipexole [59][60][61][62]. Table 1. Results of functional neuroimaging studies using dopamine, noradrenalin, serotonin, acetylcholine, inflammatory tracers, and amyloid in patients with REM sleep behaviour disorder RBD. 11 C-CFT: 2b-carbomethoxy-3b-(4trimethylstannylphenyl) tropane: 11[C]DTBZ [ 11 C]dihydro-tetrabenazine; 11 DAT dopamine transporter; DNH dorsal nigral hyperintensity; EMG electromyography; 18 F-AV133 9-18 F-fluoropropyl-(+)-dihydrotetra-benazine; 18   • iRBD patients had significantly reduced mean tracer binding in all four striatal regions at baseline (in 10 patients) and after 3 years (in 13 patients).

•
The mean reduction in tracer uptake from baseline to 3 years was 19·36% in the left putamen, 15·57% in the right putamen, 10·81% in the left caudate nucleus, and 7·14% in the right caudate nucleus.

•
The decline in tracer binding at baseline to 3 years was significantly greater in patients than in controls in all studied regions except for the right caudate nucleus. • At the 3-year assessment, three patients were diagnosed with PD (these patients had the lowest tracer uptake at baseline) • iRBD patients showed higher putamen-specific binding ratio values than PD patients with and without RBD, whereas the difference between PD groups was not significant. • PD with RBD patients showed higher caudate-specific binding ratio than patients with PD without RBD and iRBD patients. 123 I-beta-CIT-SPECT Zoetmulder et al., 2016 [27] 10/10 (+10 patients with PD and 10 with PD + RBD) • Uptake of the tracer was highest in controls, followed by iRBD patients, and lowest in PD patients.

•
When compared to controls both iRBD and PD patients with RBD showed increased EMG-activity.

•
In iRBD patients EMG-activity in the mentalis muscle was correlated with tracer uptake. • Lack of correlation between the severity of apathy and the tracer uptake in the basal ganglia.

•
The main objective of the study was the assessment of cerebral glucose metabolism and brain connectivity.

•
Decreased striatal tracer uptake in RBD patients compared with controls, and even higher decreased tracer uptake in PD and LBD patients • No significant differences in tracer uptake among the 3 study groups. 11 C-raclopride PET (striatal D 2 receptors) Wing et al., 2015 [41] 11 (with comorbid major depressive disorder)/10 (+8 with major depressive disorder without RBD) • No significant differences in the striatal tracer uptake among the 3 study groups. • Reduced tracer uptake in the primary sensorimotor cortex in the 3 study groups compared with controls. • Significant correlation between putaminal 18F-DOPA uptake and 11C-MeNER uptake in iRBD patients. • No significant differences in tracer uptake between iRBD and normal subjects at brainstem and thalamus levels, suggesting that the serotoninergic system is not involved in iRBD. 123  • Correlation between the severity of apathy in RBD patients and the tracer uptake in the dorsal raphe nuclei. • Compared with controls, patients with iRBD showed a mean 7.65% reduction in neocortical tracer.

•
The most significant reductions were found at the bilateral superior temporal cortex, occipital cortex, cingulated cortex, and dorsolateral prefrontal cortex. • Significant reduction of the tracer uptake in a global cortical region and a trend toward reduction in the substantia innominata in iRBD patients.

•
Negative correlation between a lower cortical 11C-donepezil uptake (specially in the frontal and temporal lobes) and a higher 11C(R)-PK11195 binding in the substantia innominata. • Lack of correlation between tracer uptake and iRBD duration.
Inflammation markers (microglia activation) 11  • No significant differences in tracer uptake between iRBD and controls in the substantia innominata in iRBD patients.

•
Negative correlation between a lower cortical 11C-Donepezil uptake (specially in the frontal and temporal lobes) and a higher 11C(R)-PK11195 binding in the substantia innominata.

•
Lack of correlation between tracer uptake and iRBD duration.

Noradrenaline
The possible role of noradrenaline in RBD could be suggested by the induction of this disorder by noradrenaline reuptake inhibitors such as mirtazapine [77] or duloxetine [78] (however, these drugs act as mixed serotonin-noradrenaline reuptake inhibitors), and by the beta-blocker bisoprolol [79].
Post-mortem studies have shown a severe monoaminergic cell loss in the locus ceruleus (noradrenergic) of a patient with RBD [80]. Another study did not find a significant reduction of tyrosine-hydroxylase (precursor of DA and NA) in the locus ceruleus of patients with LBD with or without concomitant RBD [81]. Schenk et al. [82] did not identify reactivity with locus ceruleus or any other brainstem area with sera from RBD patients, suggesting a lack of association of human RBD to anti-locus ceruleus antibodies. Table 1 summarizes the results of PET studies using noradrenaline transporter ligands in iRBD patients. One of them showed reduced binding of this tracer in the primary sensorimotor cortex in iRBD patients and in PD patients with and without RBD compared with controls, and a significant correlation between putaminal 18 F-fluorodihydroxyphenyl alanine ( 18 F-DOPA) uptake and 11 C-methyl-reboxetine ( 11 C-MeNER) uptake in iRBD patients [73]. Another PET study with this tracer showed a significant reduction of in the left thalamic 11 C-MeNER uptake in iIRBD patients compared with controls but similar uptake than patients with PD, while 11 C-MeNER uptake of the red nucleus was similar to that of controls and PD patients [69].
Finally, a recent study has shown lower supine norepinephrine plasma levels in iRBD patients compared with controls [96].

Serotonin
The induction of RBD by tricyclic antidepressants [97,98], selective serotonin reuptake inhibitors [97][98][99][100][101][102], or by mixed serotonin-noradrenalin reuptake inhibitors [77,78], suggests a possible role of serotonin in the pathophysiology of this disorder. Antidepressant therapy has been associated with elevated REM sleep without atonia, both in patients with and without RBD [103]. Moreover, 12.2% of subjects of a cohort of 1444 subjects under antidepressant therapy showed REM sleep without atonia in PSG studies, a significantly higher frequency than that found in 10746 subjects undergoing PSG in a sleep laboratory (2.1%; OR [95%] CI = 9.978 [8.149-12.22]) [104]. Some authors suggested that the development of RBD with antidepressant therapy should be an early signal of underlying neurodegenerative disease [102]. In contrast, an anecdotal report described improvement of RBD with selective serotonin reuptake inhibitors and worsening with the 5-HT 1A partial agonist in a single patient [105]. Table 1 summarizes the results of PET studies using serotonin transporter (SERT) ligands in iRBD patients. A study using 123 I-FP-CIT-SPECT as a marker of both DAT and SERT at the basal ganglia and of at brainstem and thalamus showed non-significant differences between iRBD patients and controls at brainstem and thalamus levels, suggesting the lack of direct implication of the serotonergic system in RBD [25]. Another study showed a correlation between the severity of apathy in RBD patients and the tracer uptake in the dorsal raphe nuclei [66]. Finally, a study using 11 C3-Amino-4-(2-dimethyl-aminome-thylphenylsulfaryl)-benzonitrile (DASB) PET, showed non-significant differences between PD patients with and without RBD in the brainstem and striatal tracer binding [46].

Acetylcholine
It has been reported induction of RBD by the acetyl-cholinesterase inhibitor rivastigmine in a patient diagnosed with Alzheimer's disease [106], while a double-blind, crossover trial involving 12 patients with PD and RBD found improvement of RBD with this drug [107], therefore suggesting a possible role of the cholinergic system in this effect. A neuropathological study showed a marked reduction in choline-acetyl-transferase labeling of neurons of the pedunculopontine/laterodorsal tegmentum nucleus in patients diagnosed with LBD, which was similar for those with or without concomitant RBD [81].
The results of brain PET studies using tracers for acetylcholine transporter and the postsynaptic acetylcholine terminal are summarized in Table 1. Patients with iRBD have shown decreased uptake of acetylcholine transporter, mainly but not exclusively, in several brainstem areas (reticular formation, pontine coeruleus/subcoeruleus complex, tegmental periaqueductal grey, and mesopontine cholinergic nuclei), which was correlated with muscle activity during REM sleep [74]. Patients with RBD associated with MSA have shown decreased acetylcholine transporter binding in the thalamus [74]. Studies on the postsynaptic acetylcholine terminal have shown reduced tracer uptake in several cortical areas in iRBD patients [68,75], and decreased neocortical, limbic cortical, and thalamic tracer uptake in PD patients with RBD compared to those without RBD [46]. Patients with RBD have decreased colonic uptake of 11 C-donepezil, suggesting loss of cholinergic gut innervation [69].
The first study with proton magnetic resonance spectroscopy ( 1 H-MRS) showed an increased choline/creatine ratio (choline is the precursor of acetylcholine) in a patient with RBD [108]. A further study involving 15 iRBD patients and 15 controls did not find significant differences between these two groups [109]. Similarly, non-significant differences were found comparing 12 patients with PD and RBD with 12 patients with PD without RBD [110]. However, a recent study involving 18 iRBD patients, 26 secondary RBD patients, and 29 controls, found a significant reduction in the same ratio in both groups of patients diagnosed with RBD [111].
Studies on the short-latency afferent inhibition (SAI) of the motor cortex with transcranial magnetic stimulation (which gives direct information about the function of some cholinergic circuits in the brain) has shown a mean reduction of this value in iRBD patients compared with controls, and a correlation between SAI and tests measuring episodic verbal memory and executive functions [112]. SAI has also been found decreased in PD patients suffering from RBD as compared both, with PD patients without RBD and with control subjects [113].
Finally, it has been described, in transgenic mice modeling brain amyloid pathologies, a significant reduction in pedunculopontine tegmentum choline-acetyl-transferase positive neurons related with impaired REM sleep [114].
2.5. Other Neurotransmitter Systems, Neuropeptides, and Hormones 2.5.1. Aspartate, Glutamate, Gamma-Amino-Hydroxybutyric Acid (GABA), and Glycine Two studies using 1 H-MRS have shown non-significant differences of N-acetyl-aspartate (NAA)/creatinine ratio in the midbrain and in the pontine tegmentum between iRBD patients and controls [109,110]. Another group found a significant decrease in NAA/Creatinine ratio in secondary RBD patients compared with iRBD and controls, and significant decrease in NAA/Choline in secondary RBD and iRBD patients compared with controls [111].
Clément et al. [115] showed in rats that the neurons of the sublaterodorsal tegmentum which triggers paradoxical REM sleep are glutamatergic and are implicated in REM sleep atonia during this sleep stage through their descending projections to medullary and spinal glycinergic premotor neurons.
The injection of adeno-associated viral vectors modifying vesicular GABA-glycine transporter (VGAT) or vesicular glutamate transporter 2 (VGLUT2) genes in mice have shown the contribution of glycinergic/GABAergic interneurons of the spinal ventral horn and glutamatergic neurons of the sublaterodorsal nucleus to REM atonia and of a separate population of glutamatergic neurons in the caudal laterodorsal tegmental and sublaterodorsal nuclei to REM sleep generation, while presynaptic GABA release in the caudal laterodorsal tegmental/sublaterodorsal nuclei, ventrolateral periaqueductal gray matter, and lateral pontine tegmentum were not involved in REM sleep control [116]. Genetic inactivation of glutamatergic neurons in the sublaterodorsal tegmentum nucleus of rats using adeno-associated viruses modifying VGLUT2 induces symptoms and behaviours during paradoxical sleep resembling human RBD [117].
Transgenic mice with deficient glycine and GABA neurotransmission show sleep, motor, and behavioral phenotype resembling clinical features of human RBD [118].
Local microinjections of glutamate in the SN of rodents bilaterally increases REM sleep, being this effect decreased by haloperidol (DA antagonist) and increased by bicuculline (GABA receptor antagonist) [119].
Injections of GABA B receptor agonists into the inferior collicollus in rats increased phasic motor activity in slow-wave sleep and tonic muscle activity in REM sleep, a finding that was associated with clinical RBD-like activity, while GABA B receptor antagonists did not induce changes [120].

Adenosine
The description of exacerbation of RBD in a patient by chocolate ingestion suggested a possible role of adenosine in this disorder [121]. It has been reported that bilateral microinjections of adenosine A 2 receptor (A 2 AR) antagonists and A 2 AR agonists into the rat olfactory bulb, and similarly inhibition and inactivation of A 2 AR, increased and decreased REM sleep respectively, and the inhibition or activation of A2AR neurons increased and decrease, respectively, REM sleep [122].

Peptides and Hormones
Hypocretin (orexin-A, a well-established marker of narcolepsy). It has been reported normality of cerebrospinal fluid (CSF) levels of this peptide in 5 patients diagnosed with iRBD [123]. However, in 63 patients diagnosed with concomitant narcolepsy and RBD, hypocretin deficiency was a predictor of symptoms of RBD [124].
Measurements of the fasting and postprandial serum levels of the orexigenic peptide ghrelin (implicated in promoting gastrointestinal motility and influencing higher brain functions) in 20 healthy controls, 39 (including 19 drug-naïve) PD patients and 11 iRBD patients have shown a less pronounced recuperation of the decrease of this peptide in the early postprandial phase in iRBD and PD patients than in controls, suggesting that ghrelin excretion should be a peripheral biomarker of both diseases [125].
A study assessing the postprandial secretion of pancreatic polypeptide and motilin involving 10 iRBD patients, 38 PD patients (19 of them drug-naïve), and 10 controls, showed a physiological pattern in all study groups, with a mild enhanced response in PD and iRBD [126].
A study assessing the 24-h blood melatonin profiles in 10 RBD patients and 10 controls showed delayed melatonin secretion by 2 h in the RBD group [127].
Finally, the serum levels of total, free, and bioavailable testosterone [128,129], and of other hormones including luteinizing hormone, follicle-stimulating hormone, estradiol-17 beta, sex-hormone binding globulin, and prolactin [129], are similar in iRBD patients than in controls.
2.6. Other Substances 2.6.1. Uric Acid Because higher plasma/serum uric acid levels have been associated with a lower risk for PD, several studies assessed plasma urate levels in patients with RBD. Plasma urate levels have been reported to be higher in patients with iRBD compared with those with PD with RBD, and a positive correlation between plasma urate levels and duration of iRBD has been found in a study [130]. However, another study found similar serum urate levels in iRBD patients and in controls, and described a longer duration of RBD and lower scores in attention, executive function, and language domains in patients with low urate levels [131].
A recent prospective study involving 12,923 Chinese adults showed an association between higher plasma urate levels and the risk for possible RBD [132].

Proinflammatory Substances
The results of studies assessing the status of markers of microglia activation in RBD patients using 1-[2-chlorophenyl]-N-[1-methyl-propyl]-3-iso-quinoline; carboxamide ( 11 C-PK11195) PET are summarized in Table 1. In summary, iRBD patients have shown decreased tracer uptake in the occipital region [68], similar tracer uptake than controls in the thalamus [68] and substantia innominata [75], and a negative correlation of higher 11 C-PK11195 binding in the substantia innominata with lower cortical 11 C-donepezil uptake, more marked in the frontal and temporal lobes [75].
Several studies assessed the serum or plasma levels of cytokines in patients with RBD. Kim et al. [133] described increased plasma levels of interleukin-10 (IL-10), and normal plasma levels of IL-1β, IL-2, IL-6, and tumor necrosis factor-α (TNF-α) in iRBD patients. Zhang et al. [134], described a significant increase of plasma TNF-α and IL-10 levels in iRBD, and decrease in plasma IL-6/IL-10 and IL-8/IL-10 ratios in iRBD patients compared with controls, and a higher predisposition to develop neurodegenerative synucleinopathies in iRBD patients with higher TNF-α/IL-10 after 3.7 years of follow-up. Finally, Kim et al. [135], found similar serum levels of IL-1β, IL-2, IL-6, IL-10, and TNF-α in iRBD patients and controls, but a higher risk of phenoconversion in patients with increased TNF-α levels and multiple markers.
A recent study showed increased blood classical monocytes and mature natural killer cells, a positive correlation of the levels of expression of Toll-like receptor 4 (TLR4) on blood monocytes in iRBD patients with nigral immune activation (assessed by 11 C-PK11195 PET), and a negative correlation with putaminal 18 F-DOPA uptake; and an opposite correlation with the percentage of Cluster of Differentiation 163 (CD163+) myeloid cells, suggesting a deleterious role for TLR4 and a protective one for the CD163 expression [136].

Alpha-Synuclein
Determinations of α-synuclein as a possible marker for prodromal PD have been performed in several tissues of iRBD and PD patients and in controls. Immunostaining of αsynuclein was similar in the colonic mucosa and submucosa of iRBD patients and controls, but that of 129-phosphorylated-α-synuclein in submucosa nerve fibers or ganglia was found in 5.3% of 19 PD patients, 23.5% of 17 iRBD patients, and in 0% of controls [137].
Aggregates of 129-phosphorylated α-synuclein in the neurofibers or the parenchyma of the submandibular gland were detected in 89% of iRBD patients, 67% of PD patients, and in 0% of controls [138]; in 44.4% of RBD patients, 46.3% of PD patients and 10.2% of controls in olfactory mucose [139]; and in skin biopsies of 86.7% iRBD patients and in no patient with narcolepsy type 1 [140]. Another study showed positivity for α-synuclein in skin biopsies of 72% of iRBD patients [141].
Finally, hypomethylation of intron 1 of the α-synuclein (SNCA) gene was found more frequently in 78 iRBD patients than in 74 controls, being hypomethylation at cytosinephosphate-guanine 17 associated with increased risk for clinical phenoconversion to neurodegenerative diseases and at cytosine-phosphate-guanine 14, 15, and 16 with disease progression [143].

Lipoprotein and Protein Glycosylation Profile
Laguna et al. [144], using 1 H NMR spectroscopy, assessed the lipoprotein profile and the presence of glycosylated proteins (a total of 27 metabolites) in the serum in 82 iRBD patients and 29 controls. Thirty-three iRBD patients showed iRBD only after 10.3 ± 4.1 years of follow-up, while 33 patients with apparent iRBD converted later to Lewy-type synucleinopathies, 35 had converted to Lewy-type synucleinopathies at the time of sample collection (20 to LBD and 15 to PD, a new sample were obtained after conversion). None of the metabolites measured differed significantly between iRBD patients who converted to Lewy-type synucleinopathies before and after this conversion and controls, but the subgroup of patients who converted to LBD showed decreased glycosylated protein B in comparison to control subjects. A comparison between samples of patients who converted to Lewy-type synucleinopathies obtained before and after this conversion showed higher levels of glycosylated protein B, and lower levels of medium sizes low-density lipoprotein (LDL) and of triglycerides in LDL after phenoconversion.

Nasal and Gut Microbiome
A recent study comparing the nasal and gut microbiome of iRBD (n = 21) and PD patients (n = 76) with controls (n = 78), assessed by 16S and 18S ribosomal RNA amplicon sequencing, showed differentially abundant gut microbes in PD and in iRBD patients, while no strong differences were found in nasal microbiota. The presence of Anaerotruncus and several Bacteroides spp. correlated with nonmotor symptoms [145]. Table 2 summarizes the results of studies assessing cerebral blood flow and brain perfusion in patients with RBD by using different methods [146][147][148][149][150][151][152][153][154][155]. Most of them showed decreased cerebral blood flow in frontal and in temporoparietal cortex [146,148,149,152,154,155], and some of them increased cerebral blood flow in the pons [148,151], periaqueductal area [151], putamen [148], hippocampus [148], and cerebellum [151]. Others have shown decreased cerebral blood flow in the parieto-occipital lobe [147,151] and in the cerebellar hemispheres [147]. Several of the described patterns were similar to those described in PD and LBD [148,149,153]. Association between brain perfusion in the frontal cortex and occipital areas with poorer performance in color discrimination test.

•
Relationship between cerebral blood flow decrease in the bilateral anterior parahippocampal gyrus and loss of olfactory discrimination • These patterns are similar to those seen in PD and LBD. • Follow-up study of 9 patients of cohort described in reference [22] 8 months later [149]. • Decrease in regional cerebral blood flow in bilateral parietotemporal and occipital areas in the two studies.  • Patients with iRBD showed decreased regional perfusion in the anterior frontal and lateral parietotemporal cortex compared with controls. • A second SPECT performed 17 months later in iRBD patients showed a relative increase (with reversion to normal levels in controls) in regional perfusion in the anterior frontal, lateral parietal, and occipitotemporal cortices Dynamic susceptibility contrast MRI using spin echo echo-planar imaging with high resolution T1-weighted images using a 3.0 Tesla MRI unit • Patients with iRBD showed profound hypoperfusion and microvascular flow disturbances throughout the cortex in patients compared to controls, especially in cortical areas related to language comprehension, visual processing, and recognition.

•
Cortical hypoperfusion was associated with impaired cognitive performance.
According to several studies, alterations in brain perfusion were related to color discrimination tests [148,150], loss of olfactory discrimination [148], and global cognitive performance [149,155].

•
The previously described alterations decreased with the progression of the disease. • iRBD subjects showed higher PDRP expression than controls, but lower than that of PD and LBD. • iRBD subjects with hyposmia and/or abnormal DAT scan showed higher PDRP expression • PDRBDRP showed a relative increased glucose metabolism in the premotor cortex and hippocampus, and also reflected the PD-related covariance pattern. reported previously • PDRBDRP was significantly higher in PDRBD patients than in iRBD patients, and in iRBD patients than in controls.

•
Negative correlation of PDRBDRP with olfactory and frontal executive functions and with the striatal DAT density.

Authors, Year [Ref] Patients/Controls Method Main Findings
Liguori et al. [161] 54/35 (+28 patients with PD, 10 with LBD and 55 with Alzheimer disease -AD) 18 F-FDG PET • iRBD patients showed increased 18F-FDG uptake in the brainstem and in several areas of the temporal lobe, limbic lobe and frontal lobe, and reduced 18F-FDG uptake in temporal and parietal regions compared to controls. • iRBD patients showed several differences in 18F-FDG uptake when compared with PD (increased uptake in the midbrain and pons, cerebellum, lentiform nucleus, claustrum, and frontal and temporal lobes), LBD (relative hypometabolism in frontal and limbic lobes), or AD groups, being the comparison with AD patients that which showed the main differences (reduced uptake in the frontal lobe and increased uptake in limbic, temporal and parietal lobes). • Future phenoconversion of iRBD patients was predicted, with all cut-off ranges, by high dnPDRBDRP.

•
The predictability of PDRP future phenoconversion only was significant in a partial range of cut-off.
The so-called PD-related pattern (PDRP), characterized by decreased cortical metabolic activity in the temporal, lateral occipital, inferior and posterior parietal, and prefrontal association cortices and in the supplementary motor area; and increased metabolism in the cerebellum and pons, pallidum, thalamus, paracentral lobule, limbic and sensorimotor cortices, and in the left supplementary motor area, was found with higher frequency in iRBD patients than in controls, but with lower frequency than in PD and LBD patients [29,159], while the iRBD-related pattern (iRBDRP, relative increased glucose metabolism in the brain stem, thalamus, cerebellum, hippocampus, and sensorimotor cortex, and decreased metabolism in the middle cingulated, parietal, temporal, and occipital cortices) was similar in iRBD and in PD patients and showed higher expression in both PD and iRBD patients than in controls [159]. A mixed PD and iRBD related pattern (PDRBDRP), which was more marked in PD patients than in iRBD patients and in both groups than in controls, has also been described [45]. In patients with iRBD and decreased DAT uptake and in patients with PD, but not in patients with iRBD and normal DAT uptake, DAT binding was correlated with PDRP expression [45]. The presence of de novo PDRBDRP (dnPDRBDRP) in iRBD patients seems to be a predictive marker for future phenoconversion [162]. It has also been described, both for iRBD and PD patients, increased glucose metabolism in the pons, thalamus, medial frontal and sensorimotor areas, hippocampus, supramarginal and inferior temporal gyri, and posterior cerebellum [156,157,161], and normal 18 F-FDG uptake in the striatum of RBD patients [41].

Structural and Functional Magnetic Resonance Imaging (Mri) Studies
The results of studies assessing structural and functional MRI (fMRI) studies are summarized, in chronological order, in Table 13  . These include sudies of gray matter volume (GMV), cortical thickness, diffusion tensor imaging, dorsal nigral hyperintensity (DNH), neuromelanin-sensitive structural and diffusion MRI, and functional MRI. Table 4. Results of structural and functional neuroimaging in patients with REM sleep behavior disorder (RBD). AD axial diffusivity; ALFF amplitude of low-frequency fluctuations; BG basal ganglia; BOLD blood oxygen level-dependent; BS brainstem; CSO centrum semiovale; DAT dopamine transporter; DBM deformation-based morphometry; DFC dynamic functional connectivity; DNH dorsal nigral hyperintensity; DTI diffusion tensor imaging; EPVS enlarged perivascular space; FA fractional anisotropy; fMRI functional MRI; GE-EPI gradient echo planar imaging; GM gray matter; GMD gray matter density; GMV gray matter volume; iRBD idiopathic RBD; MCI mild cognitive impairment; MD mean diffusivity; MRI magnetic resonance imaging; MVPA multivariate pattern analysis; NM neuromelanin; PD Parkinson's disease; RBD REM sleep behavior disorder: RD radial diffusivity; ReHo regional homogeneity; SN substantia nigra; SWI susceptibilityweighted imaging; T Tesla; TIV total intracranial volume; VBM voxel-based morphometry.

RBD/ Controls Method Main Findings
Unger et al., 2010 [163] 12/10 1.5 T MRI single-shot echo planar sequence with a twice-refocused spin echo pulse, sequence acquisition, processed with voxel-wise analysis of DTI. Measurement of AD (a potential marker of neuronal loss), RD (a potential marker of glial pathology), and FA (measure of brain-tissue integrity)

•
The white matter of the brainstem, the right substantia nigra, the left temporal lobe, the olfactory region, the fornix, the corona radiata, the internal capsule, and the right visual stream, showed significant microstructural changes in patients with iRBD. • GMV in the tegmental portion of the pons, anterior lobes of the both cerebellar hemispheres, and left parahippocampal gyrus was significantly reduced in patients with iRBD

•
At least two-thirds of iRBD patients showed (this frequency approaches to the rate observed in PD and higher than that found in controls).

•
According to the authors the absence of DNH could be a useful tool to identify prodromal degenerative parkinsonism in patients with iRBD.

•
These results suggest an association of RBD with loss of volume in the pontomesencephalic tegmentum (where cholinergic, GABAergic, and glutamatergic neurons coexist). • Cortical thickness analysis showed thinning in iRBD patients in the anterior cingulate orbitofrontal bilateral medial superior and frontal cortices of both hemispheres, and in the right dorsolateral primary motor cortex. • iRBD patients showed decreased GMV in the anterior cingulate gyri, frontal lobes, and caudate nucleus. • Compared with controls, iRBD patients showed a reduction in the NM-sensitive volume and signal intensity and a decrease in FA, and no differences in AD, RD, or MD or in R2*.

•
The receiver operating characteristic analysis of NM-sensitive volume and signal intensity was able to discriminate between iRBD patients and controls with a high diagnostic accuracy. • These three biomarkers had a combined accuracy of 0.92. • iRBD patients with depressive symptoms showed reduced GMV in the caudate nucleus compared to controls and iRBD patients without depressive symptoms. • iRBD patients with anxiety symptoms showed reduced GMV in the left amygdala extending to the hippocampus compared to controls and iRBD patients without anxiety symptoms.

•
Higher scores for depression and anxiety in iRBD patients were associated with lower GMV in these regions respectively. 3.0 T MRI T1-weighted acquisition resting state whole-brain fMRI acquired with single-shot gradient echo planar imaging (GE-EPI).
• When compared to iRBD patients with normal motor function and with controls, iRBD patients with mild motor impairment showed increased cortico-cerebellar functional connectivity and decreased cortico-striatal functional connectivity Table 9. Cont.

RBD/ Controls Method Main Findings
Ohlhauser et al., 2019 [179] 17 (all converted further to PD)/21 (+20 prodromal PD patients, 14 with RBD and 6 with hyposmia) 3.0 T MRI T1-weighted acquisition, and processed with DTI. Measurement of MD and FA by using tract-based spatial statistics • Significant increased in MD in prodromal PD relative to iRBD in widespread, but mostly right-lateralized regions (this pattern was particular to individuals with RBD). • Lack of microstructural differences between controls, prodromal PD, and iRBD patients.

•
The prodromal PD group had significantly higher RBD symptoms and the iRBD group had significantly higher motor symptoms. 3.0 T MRI T1-weighted acquisition; study of brain functional connectivity using resting-state fMRI • Compared with controls, iRBD patients showed reduced cortico-cortical functional connectivity strength in edges located in posterior regions (this regional pattern was also shown in an independent analysis comprising posterior areas where a decreased connectivity in 51 edges was found, whereas no significant results were detected in the anterior network). 3.0 T MRI T1-weighted acquisition; resting-state fMRI and seed-to voxel analysis were used to study thalamo-cortical functional connectivity • iRBD patients showed higher functional connectivity than controls between the left thalamus and occipital regions (these included the left fusiform gyrus, and lingual gyrus, and the right cuneal cortex). • Thalamo-fusiform functional connectivity was positively correlated with word list recognition • iRBD patients and controls showed significant differences in cortical thinning progression in bilateral precuneus and superior parietal areas, in the left occipital pole and lateral orbitofrontal gyri, and in the right cuneus • Worse progression in the visual form discrimination test in iRBD patients was associated with loss of gray matter in the left precuneus and in the right superior parietal area. • Cortical thinning (specially in frontal regions) was associated with increased motor signs in iRBD patients.

•
Cortical thinning involving posterior areas was associated with late-onset iRBD.   3.0 T MRI T1-weighted acquisition, processing with SWI and quantitative susceptibility mapping for evaluation of the nigrosome-1 (N1) sign in the SN, global and regional high-iron content, and volume of subcortical nuclei.
• Compared with controls, a higher number of iRBD patients showed N1 loss. • iRBD patients showed a reduced volume of the right caudate nucleus.

•
Global and RII iron of the subcortical nuclei was similar in iRBD patients and controls

Studies on Gray Matter Volume (GMV)
Most studies on GMV have used voxel-based morphometry. Although several of these studies have not shown significant clusters of reduced GMV or similar total brain volumes in between iRBD patients compared with controls [164,178,189], others have shown nonhomogeneous results in iRBD patients, including smaller bilateral putamen volumes [164], decreased GMV in the anterior lobes of the right and left cerebellum, tegmental portion of the pons, and left parahippocampal gyrus [166], decreased GMV in the frontal lobes, anterior cingulate gyri, and caudate nucleus [172], decreased GMV in the frontal, temporal, parietal, occipital cortices as well as increased GMV in cerebellum posterior lobe, putamen, and thalamus [182], decreased GMV in the brainstem, anterior cingulate and insula [191], decreased GMV of the right caudate nucleus [194], and right hippocampal atrophy [177]. Bourgouin et al., 2019 [176] described reduced GMV in the caudate nucleus of iRBD patients with depressive symptoms and in the left amygdala-hippocampus in iRBD patients with anxiety symptoms when compared to iRBD patients without these symptoms and with controls. In contrast, other studies have shown increases of GMV in both hippocampi [165], or the frontal cortex, thalamus, and caudate nucleus of iRBD patients [175].
Compared with PD patients without RBD and controls, patients with concomitant PD and RBD or probable RBD have shown decreased volumes in the pontomesencephalic tegmentum (location of cholinergic, GABAergic, and glutamatergic neurons), medullary reticular formation, hypothalamus, thalamus, putamen, amygdala, and anterior cingulate cortex [171], volume decrease over time in the left caudate nucleus, pallidum and amygdale [187], and a relatively high GMV in the cerebellar vermis IV/V and low GMV in the right superior occipital gyrus [188].

Studies on Cortical Thickness
Compared with controls, iRBD patients showed, according to Rahayel et al. [172] thinning in bilateral anterior cingulate, orbitofrontal, and medial superior frontal cortices, and in the right dorsolateral primary motor cortex. Patients without MCI showed cortical thinning in the frontal cortex only, while those with cognitive impairment showed cortical thinning in the cingulate, temporal, frontal, and occipital cortices, and abnormal surface contraction in the thalamus and in the lenticular nucleus [174].
Campabadal et al. [177] described cortical thinning in the left superior parietal, postcentral, and fusiform regions, and right superior frontal and lateral occipital regions, and significant differences between iRBD patients and controls in the progression of cortical thinning in the left occipital pole and lateral orbitofrontal gyri the right cuneus, and bilateral superior parietal lobe and precuneus [184]. Pereira et al. [178] reported parietal and occipital cortical thinning in iRBD patients compared to controls, and cortical thinning in frontal, occipital, and parietal areas in iRBD patients who developed further neurodegenerative disease compared to iRBD non-converters.
Compared with PD patients without RBD, patients with PD and RBD or probable RBD showed bilateral inferior temporal cortex thinning, and an increase in the rate of thinning in the left insula during a follow-up period [187].

Studies Using Diffusion Tensor Imaging (DTI)
While one study showed significant microstructural changes in the white matter at several areas including the brainstem, the right substantia nigra, the left temporal lobe, the olfactory region the fornix, the corona radiata, the internal capsule and the right visual stream of iRBD patients compared with controls [163], other failed to find any microstructural difference between iRBD patients, prodromal PD patients, and controls, although described a significant increase in mean diffusivity (MD) in prodromal PD relative to iRBD in widespread, but mostly right-lateralized regions [179]. A study described decreased fractional anisotropy (FA) in the rostral pons and in the tegmentum of the midbrain and increased MD within the pontine reticular formation overlapping with a cluster of decreased FA in the midbrain in iRBD patients [90], and other decreased in FA, and no differences in axial diffusivity (AD), radial diffusivity (RD), or MD in the SN in iRBD patients compared with controls [173].
Finally, a study of free-water maps with a bi-tensor model showed higher free-water values in the posterior SN in patients with iRBD (which increased over time) than in controls, but significantly lower than in patients with PD [193].

Studies Addressing Dorsolateral Nigral Hyperintensity (DNH)
Two studies addressed the measurement of DNH by using high-field susceptibilityweighted imaging (SWI). One of them described that at least two-thirds of iRBD individuals showed loss of DNH, a frequency which was similar to that observed in PD and higher than that found in controls [170], while other showed loss of DNH in only 27.5% of RBD patients (and 7.7% of control subjects and 96% of PD patients) [35]. It is possible that loss of DNH could be useful in identify prodromal degenerative parkinsonism in iRBD.

Studies with Neuromelanin-Sensitive Structural and Diffusion MRI
Studies using this method showed reduced signal intensity in the locus coeruleus/subcoeruleus complex [169] and in the SN [173], in iRBD patients compared with controls. PD patients also have shown reduced signal intensity in the locus coeruleus/subcoeruleus area compared with controls, which was more marked in patients with coexistent RBD [167].

Studies with Functional MRI (fMRI)
The first study with resting-state fMRI (rsfMRI) showed decreased nigrostriatal and nigrocortical connectivity (between SN and right cuneus/precuneus and superior occipital gyrus) in iRBD patients compared with controls but increased connectivity compared with PD patients [168]. Further studies have shown the following findings:

1.
Decreased functional connectivity in the basal ganglia network both in iRBD and in PD patients which differentiated iRBD and PD from controls with high sensitivity and specificity, but did not differentiate RBD from controls [28].

2.
Reduced cortico-cortical functional connectivity strength in edges located in posterior regions in iRBD patients compared with controls [180].

3.
Increased functional connectivity between the left thalamus and occipital regions including the right cuneal cortex, left fusiform gyrus, and lingual gyrus in iRBD patients compared with controls [181].

4.
Decreased functional connectivity between the limbic striatum and temporo-occipital regions in iRBD patients compared with controls, which was associated with the presence of impulse control disorder in iRBD patients [189].

5.
Decreased functional connectivity between the brainstem and the anterior cingulated, temporal lobe, and the cerebellum posterior lobe in iRBD patients compared with controls [191]. 6.
Decreased striatal-prefrontal (this in executive control) and midbrain-pallidum functional connectivity in the basal ganglia network, decreased motor and somatosensory cortex functional connectivity in the sensorimotor network, and lack of abnormalities in the default mode network compared with controls [190]. 7.
Decreased functional connectivity with two clusters located in the precuneus in iRBD patients compared to the controls using a multivariate pattern analysis [186].
A study on structural connectivity showed decreased measures of the global network, average degree, global efficiency, and local efficiency, and increased measures of characteristic path length in iRBD patients compared to controls [175]. Two studies measuring amplitude of low-frequency fluctuations (ALFF), showed, respectively, increased ALFF values in the right parahippocampal gyrus [182], and normal values [36]. A significant reduction in regional homogeneity in the bilateral putamen of iRBD patients compared with controls has also been described [36].
Finally, two studies addressed the possible differences in functional connectivity (one of them in dynamic functional connectivity) between PD patients with and without concomitant RBD. One of them showed decreased functional connectivity between the right superior occipital gyrus and the posterior regions (left fusiform gyrus, left calcarine sulcus, and left superior parietal gyrus [188], and the other a brain pattern mainly marked by weaker positive couplings between visual network and default mode network, default mode network and basal ganglia network, and within default mode network [192], in PD patients with RBD compared with those without RBD.

Measurement of Visible Enlarged Perivascular Space (EPVS)
A recent study described higher burdens of EPVS in the centrum semiovale, basal ganglia, substantia nigra, and brainstem of iRBD patients in comparison with PD patients and controls, being higher EPVS in the centrum semiovale and the brainstem as independent risk factors for iRBD [185].

Conclusions
There are enough data suggesting the role of dopaminergic dysfunction in iRBD. Data from functional neuroimaging studies with PET/SPECT have shown that many patients diagnosed, at least initially, with iRBD, had decreased uptake of tracers for the presynaptic dopaminergic terminal, suggesting the presence of preclinical parkinsonism, and, moreover, many of the patients develop PD or other synucleopathies later. The high frequency of SNH in patients with iRBD is consistent with these data, and the combination of functional neuroimaging studies with TCS has shown to be useful to predict the phenoconversion of iRBD to a synucleopathy [21].
Data from functional neuroimaging studies showing a decreased uptake of the acetylcholine transporter in the brainstem [74] and decreased uptake of tracers of the postsynaptic cholinergic terminals in several cortical areas [68,75], and in the colon [69], suggest a possible role of acetylcholine in the pathophysiology of iRBD. In addition, the results of studies on SAI of the motor cortex are consistent with the presence of a cholinergic deficit in patients with iRBD and with RBD associated with PD [112,113], giving support to this hypothesis. In contrast, the results of measurements of choline/creatinine ratio in 1 H-MRS studies have not shown conclusive results.
The peptide ghreline, the hormone melatonin, proinflammatory markers such as IL-10 and TNF-α, several lipoproteins and glycosylated proteins, and specially alpha-synuclein, could be considered as potentially interesting markers for RBD and to predict its possible phenoconversion to synucleopathies.
In general, brain perfusion studies and brain glucose metabolism studies have shown increased cerebral blood flow and glucose metabolism in the pons, thalamus, hippocampus, and cerebellum, decreased cerebral blood flow in the temporoparietal cortex and decreased cortical metabolic activity with predominance in temporoparietal, prefrontal, occipital, and supplementary motor area, a pattern resembling that of PD and LBD.
The results of structural and fMRI are still controversial (Table 13). Most of them have shown the presence of structural changes in deep gray matter nuclei, cortical gray matter atrophy, and alterations in the functional connectivity within the basal ganglia, the cortico-striatal, and the cortico-cortical networks [195]. However, there is a lack of homogeneity of both the methods used and the results obtained, and many of these studies were based on low sample sizes.
The design of prospective multicenter studies involving a large series of iRBD patients and controls with a long-term follow-up period, using multiple biochemical and multimodal neuroimaging parameters should be appropriate in an attempt to establish the neurochemical features of iRBD and the risk factors for its phenoconversion to synucleopathies. Funding: The work at the authors' laboratory is supported in part by Grants PI15/00303, PI18/00540, and RETICS RD16/0006/0004 from Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Spain, and IB16170, GR18145 from Junta de Extremadura, Spain. Financed in part with FEDER funds from the European Union.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.