REM sleep behavior disorder (RBD) is commonly observed in several neurodegenerative diseases, especially in synucleinopathy, which includes Parkinson’s Disease (PD). A total of 30%–60% of patients with PD present with RBD during their disease course; patients with RBD seem to be a distinct clinical entity compared with patients with PD without RBD [1, 2]. Previous cross-sectional studies have reported poorer motor features, cognitive dysfunction, and hallucination in PD with RBD compared with PD without RBD, and a longitudinal study has reported poorer clinical outcomes with RBD [3, 4, 5]. However, it is unknown why PD with RBD showed distinct clinical phenotype.

Neuronal damage causing parkinsonism is mainly on the degeneration of nigrostriatal pathway in PD. In addition, striatum has wide connection with various anatomical structures including cortex and other subcortical areas, hence, subregional difference in striatal degeneration can contribute to clinical phenotype of PD. Considering PD with RBD have a distinct clinical entity compared to PD without RBD, there is a possibility of subregional difference in the striatal degeneration associated with RBD. Nonetheless, the denervation pattern in the striatal subregions has never been evaluated according to the existence of RBD in PD. In this study, we aimed to investigate whether the nigrostriatal degeneration pattern differs between patients with RBD and those without RBD (non-RBD) with early-stage PD who are naïve to drug therapy.

Table 1 shows the different demographic and clinical features of the RBD and non-RBD groups. Age, gender, and clinical features related to parkinsonism (disease duration, akineto-rigid phenotype, Hoehn and Yahr’s scale, and Unified Parkinson Disease Rating Scale) were not significantly different between the two groups (p $\mathrm{>}$ 0.5 for all). Global cognition, assessed using the Montreal Cognitive Assessment test, did not differ significantly between the groups (p = 0.254).

Among the six striatal areas, the PP area showed the lowest SBR values in both RBD and non-RBD groups and in all participants (p 0.001, analysis of variance). Comparing the other areas, the VS area had lower SBR in the RBD group than in the non-RBD group in the less and more affected side of the striatum (p = 0.032 and p = 0.026, respectively). The bilateral sum SBR values of the AC and VS areas had lower SBR in the RBD group than in the non-RBD group (p = 0.046 and p = 0.020, respectively). However, this statistical significance was not definite after multiple comparison correction (p $>$ 0.05 in all). In the other areas (PC, AP, PP, and VP) there was no significant difference in SBR between the two groups even without multiple comparison corrections (Table 2, Fig. 1).

Fig. 1.

Tophography of 6 striatal areas and the areas showing different value according to the existence of RBD in PD. AC and VS area showed lower striatal binding in the RBD group compared to non-RBD group.

Comparing the left and right sides of the striatum, the left VS had significantly lower SBR in the RBD group than in the non-RBD group (p = 0.017); the right VS showed a moderate difference (p = 0.051). In addition, the right AC had lower SBR in the RBD group than in the non-RBD group (p = 0.047) (Supplementary Table 1). However, this statistical significance was not definite after multiple comparison correction (p $>$ 0.05 in all).

In cognitive assessment, the forward digit span in attention domain and the Controlled Oral Word Association Test (animal) assessment of frontal/executive function were decreased in the RBD group compared with the non-RBD group (p = 0.35 and p = 0.32, respectively). Across the five cognitive domains, the RBD group had a significantly decreased attention function compared with the non-RBD group (p = 0.022). There were no statistically significant differences between the groups in the other cognitive domains (language, visuospatial memory, and frontal function) (Table 3). Dysfunction in attention domain was well correlated with SBR in VS (r = 0.358, p 0.05, Supplementary Table 2).

We evaluated subregional differences in presynaptic dopaminergic neurodegeneration of the striatum with regard to RBD in drug-naïve patients with early PD. Overall, striatal SBR was significantly decreased in the PP area, as expected, irrespective of RBD; however, in the AC and VS areas, which are relatively well preserved in early PD, the RBD group had a significantly lower uptake than the non-RBD group in this study.

Striatal degeneration in PD shows a distinctive spatial and temporal pattern during the course of the disease. In PD, a degeneration pattern with an anteroposterior gradient has been reported in previous literature [9]. This pattern is quite different from the Parkinson Plus syndrome [9]. From the early stage of PD, the PP area is the most affected; this corresponds well with the pattern of neurodegeneration in the substantia nigra, i.e., early degeneration of the ventrolateral substantia nigra, which was demonstrated in autopsy studies [10, 11]. This pattern was common in all our study patients irrespective of RBD. The severity of degeneration in the PP area did not differ between the RBD and non-RBD groups. However, among the relatively preserved areas of the striatum, uptake in the AC and VS differed significantly between the groups according to the presence or absence of RBD.

The PP area has a primary connection with the motor cortical area, which may explain why motor dysfunction is prominent in early PD [12]. The VS area may have strong connectivity with the ventromedial prefrontal cortex and orbitofrontal cortex, which may imply that it has limbic and attentional functions [12, 13]. Similarly, the AC area is connected to the prefrontal cortex and the major parts of the orbitofrontal cortex or dorsal anterior cingulate cortex [13]. From these connection, the VS and AC areas are associated with emotional and cognitive function. Although we did not evaluate reward or emotional aspects sufficiently, our cognitive assessments indicated that attentional deficit was prominent in the RBD group; this dysfunction may be associated with the different neurodegeneration levels observed in the AC or VS area or both.

There have been few positron emission tomography imaging studies comparing the patterns of nigrostriatal dopaminergic degeneration in PD between individuals with and without RBD. Arnaldi et al. [14, 15] used I${}^{123}$-FP-CIT–single-photon emission computerized tomography (SPECT) in patients with PD with and without RBD. They found that patients with PD and RBD had worse cognitive function and more severe nigrostriatal dopaminergic impairment in the caudate area of the less affected hemisphere than those without RBD [14, 15]. Our previous study using I${}^{123}$-FP-CIT-SPECT of 416 patients with de novo PD (from the Parkinson’s Progression Markers Initiative study cohort) found that dopamine transporter uptake did not differ between RBD and non-RBD groups at baseline; however, during 4 years of follow-up, striatal SBR was found to decrease faster in the RBD group that in the non-RBD group. Furthermore, the caudate area showed greater denervation, compared with the putamen area [16]. However, previous studies have been limited by the low resolution of SPECT imaging, making it difficult to distinguish the striatal subregions. SPECT imaging can analyze striatal dopaminergic density only in two regions, the caudate and putamen. Our results show low density in the AC and VS areas in the RBD group, which is markedly different from previous findings. Our study clarifies that the ill-defined area named “caudate” in SPECT imaging can be divided into the AC and VS subregions.

Although our study has a limitation in that multiple test corrections by Bonferroni did not find any significant statistical differences, this study is an exploratory study design and has a small sample size. However, this study did not demonstrate the statistical difference in multiple comparisons, it should be evaluated in larger scale or longitudinal design. Therefore, based on the current study, further study is required for the interpretation of these findings. Nevertheless, we used polysomnography to confirm RBD, unlike previous studies that used clinical diagnoses to establish RBD. Although we included only a small number of patients, this study has methodological merits because RBD confirmed by polysomnography increased sensitivity for determining RBD-related characteristics.

The regional selectivity of nigrostriatal degeneration may be observed in PD with RBD compared with PD without RBD implies that there are different clinical phenotypic presentations. It remains unknown why this differential degeneration pattern occurs in PD. Thus, further investigation using large scale is required for statistical confirmation.

IK, investigation and data curation, original draft preparation. YKL, investigation and data curation, original draft preparation. HM, and YJK, investigation and data curation. SL, investigation and data curation, review and editing of manuscript. MY, investigation and data curation, review and editing of manuscript. HSH, Investigation, review and final preparation of review and editing of manuscript. YEK, investigation, data curation, original draft preparation, review and final preparation of review and editing of manuscript.

This study was approved by Hallym University Sacred Heart Hospital Institution’s ethical committee (IRB 2019-06-008). Informed consent was waived by IRB because medical records were used only for this study.

The authors thank Jaeseol Park for advice on neuropsychological test.

This research was supported by Hallym University Research Fund and the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (2020R1F1A1076697).

The authors declare no conflict of interest.