A bibliometric analysis of RBD-related PD publications was performed using the Web of Science Core Collection (WoSCC) and Scopus databases, both recognized for their authoritative and comprehensive coverage of globally influential academic literature. To maximize search accuracy and completeness, keywords were initially compiled from a preliminary literature review and subsequently refined using PubMed MeSH terms. The search was conducted on September 1, 2025, encompassing all records from the inception of each database up to the specified date. The final search query in the WoSCC was formulated as: TS = (“REM Sleep Behavior Disorder*” OR “REM Behavior Disorder*” OR “Rapid Eye Movement Sleep Behavior Disorder*”) AND TS = (parkinson*). In WoSCC, 2,312 publications were initially identified, with subsequent filtering to include only “Article” and “Review Article” document types while excluding retracted publications, resulting in a final dataset of 1,916 articles (Figure 1). In Scopus, the search was performed using “Article title, Abstract, Keywords,” the search query was formulated as: TITLE-ABS-KEY (“REM Sleep Behavior Disorder*” OR “REM Behavior Disorder*” OR “Rapid Eye Movement Sleep Behavior Disorder*”) AND TITLE-ABS-KEY (parkinson*). In Scopus, 2,921 records were initially identified, and after restricting to “Article” and “Review” types 2,536 articles were retained. The WoSCC search results were exported in Plain Text File format, including Full Records and Cited References, while the Scopus records were exported in “CSV with All Fields.” Records retrieved from the WoSCC and Scopus were combined and managed using EndNote 2025. Duplicate records were removed through a two-step procedure. First, EndNote’s automatic duplicate detection function was applied to identify records with identical Digital Object Identifiers (DOIs). Second, a manual screening was performed to identify residual duplicates based on article titles, author(s), and publication years. When duplicate records were identified across databases, the WoSCC version was retained, as its metadata are generally more standardized and more compatible with bibliometric analysis software. This approach ensured the accuracy and consistency of the final dataset, resulting in 2,887 unique publications for subsequent analysis.

To ensure the reliability of the data, two researchers independently conducted the literature search, document download, and bibliometric data extraction on the same day. Any discrepancies were resolved through consensus discussions with a third researcher.

CiteSpace was employed to consolidate keyword synonyms, and remove irrelevant terms, ensuring clarity and consistency in subsequent keyword analyses. Following this, bibliometric and knowledge mapping analyses were conducted using VOSviewer 1.6.20, Bibliometrix, CiteSpace 6.3. R1, R software, and Microsoft Excel 2019.

Analyzing the annual publication volume and trends in a research field provides insights into its academic impact and global research dynamics (Chen and Song, 2019). As of September 1, 2025, the WoSCC and Scopus databases indexed 2,887 publications on RBD in PD, with a total of 76,730 citations, yielding an average of 41.49 citations per article and an H-index of 117. The earliest publication appeared in 1993, with the highest publication volume recorded in 2024, totaling 301 articles. The annual publication growth rate has averaged 17.17%, with a correlation coefficient (R²) of 0.87, indicating a strong upward trend. Notably, the growth rate has accelerated significantly since 2010 (Figure 2), with three major peaks in publication volume observed in 2013, 2021, and 2024. Citation activity began to rise more sharply from 2018 onward, following an almost exponential trajectory, with citation peaks occurring in 2013, 2017, and 2022.

Country analysis highlights the leading contributors in RBD-related PD research and evaluates patterns of international collaboration. Research output spans 61 countries, with a country centrality analysis conducted using CiteSpace (Figure 3A). Centrality values indicate a country’s pivotal role within the field, with six nations exhibiting high centrality: the USA (0.35), Japan (0.27), England (0.25), Germany (0.19), Canada (0.16), and France (0.14), signifying their leadership roles in advancing RBD-related PD research.

The country-specific publication time-zone chart illustrates the initiation year and annual publication trends for each country (Figure 3B). The top ten countries by publication volume are the USA, China, Italy, Canada, Germany, France, Japan, England, Spain, and Korea. The USA and Japan were the earliest to contribute to the field, whereas China’s involvement began in 2008, with a markedly rapid increase in publication rate in recent years (Figure 3C).

Country Citation Burst Analysis assesses research activity and impact in the RBD-PD domain (Figure 3D). The earliest burst began in 1998, with Luxembourg exhibiting the longest duration (1998–2014), while the most intense citation burst is observed in China (15.57), which continues. Inter-country collaboration analysis was performed for the 30 countries with more than eight co-authored publications (Figure 3E). The USA emerged as the most collaborative, demonstrating strong research partnerships with Germany, Italy, France, and Canada.

Institutional analysis identifies key institutions in RBD-related PD research and examines collaboration patterns, offering insights for enhancing future cooperation and knowledge exchange. A total of 1,889 institutions have contributed to this field, with the Hennepin County Medical Center from the USA exhibiting a high centrality score (0.10), underscoring its pivotal role (Figure 4A). It is also one of the earliest institutions to engage in this research and has maintained a continuous presence.

Prominent institutions leading in publication volume include McGill University, Université de Montréal, Mayo Clinic, University of Barcelona, and Philipps University Marburg (Figure 4B). Analysis of the Top 25 Cited Institutions with the Strongest Citation Bursts (Figure 4C) shows that six institutions continue to experience significant citation activity. These active institutions are the University of Cagliari and Università di Modena e Reggio Emilia (Italy), Capital Medical University and Zhejiang University (China), and the University of Cologne and Research Center Jülich (Germany), indicating ongoing influential research efforts.

Institutional collaboration analysis was conducted for the 28 institutions with over 25 co-authored publications (Figure 4D). The strongest collaborative relationships were observed between Capital Medical University and Shanghai Jiao Tong University (China), and between McGill University and Université de Montréal (Canada). This indicates a need for expanding international collaboration to further advance research in this domain.

Author analysis provides insights into the leading experts in RBD-related PD research, offering a reference point for potential collaborations. A total of 6,692 authors have contributed to the field, with an international collaboration rate of 26.85%. The top 10 most prolific authors include Gagnon Jean-François, Postuma Ronald, and Montplaisir Jacques Y from Canada; Iranzo Alejandro and Santamaria Joan from Spain; Boeve Bradley F from the USA; Hoegl Birgit from Austria; Arnulf Isabelle from France; Šonka Karel from the Czech Republic; and Plazzi Giuseppe from Italy. Most of these researchers began focusing on this area between 2006 and 2009 and have continued to engage in substantial, ongoing research (Figures 5A,B).

The Top 25 Cited Authors with the Strongest Citation Bursts analysis identified Frauscher Birgit as having the strongest burst strength, while Montplaisir Jacques Y exhibited the longest citation burst duration (Figure 5C). Several authors showed ongoing citation bursts, indicating continued influence. A co-authorship analysis was conducted on 110 authors with more than 10 co-occurrences (Figure 5D). These authors were grouped into 11 clusters based on collaboration intensity and research focus. Figure 5E illustrates the key authors within each research direction, highlighting their roles in advancing RBD-related PD studies.

Journal analysis identifies key journals in RBD-related PD research, aiding researchers in tracking trends, staying informed, and selecting suitable publication venues. A total of 297 journals have published work in this area. The yearly publication output of the top 10 journals is shown in Figure 6A. According to Bradford’s Law Core Sources (Figure 6B), the most influential journals are Movement Disorders, Sleep Medicine, Parkinsonism Related Disorders, Neurology, and Sleep, which together account for 35.34% of the total publications (1,729). Notably, four of these journals are ranked in the Journal Citation Reports (JCR) Q1 category, reflecting their high research quality.

The journal overlay visualization (Figure 6C) illustrates the developmental pathways and trends in RBD-related PD research. Given the database integration constraints, the dual-map overlay of journals was generated using WoSCC data. The Citing Journals primarily from fields such as medicine, clinical sciences, molecular biology, immunology, neurology, and sports science. The Cited Journals mainly belong to psychology, education, social sciences, molecular biology, and genetics. The three thickest citation paths indicate a shift in research focus from molecular biology and psychosocial education toward molecular immunology, clinical medicine, and neurology-related sports sciences. Citation burst analysis identified journals with strong historical and ongoing influence, including Frontiers in Aging Neuroscience and NPJ Parkinson’s disease (Figure 6D).

Reference analysis uncovers the knowledge structure, research hotspots, interdisciplinary connections, and academic influence within a field, offering insights into its historical and current status while guiding future research directions and strategic decisions (Chen et al., 2010). In the domain of RBD-related PD research, the Top 10 Highly Cited Papers (Supplementary Table S1; Wirdefeldt et al., 2011; Stiasny-Kolster et al., 2007; Seppi et al., 2011; Seppi et al., 2019; Schenck et al., 1996; Schenck et al., 2013; McKeith et al., 2005; McKeith et al., 2017; Connolly and Lang, 2014; Armstrong and Okun, 2020) primarily focused on RBD diagnosis, management, and treatment of RBD, as well as its association with PD and dementia with Lewy bodies (DLB). These studies underscore the role of RBD as a precursor symptom for PD and DLB, highlighting the significance of early detection and intervention in mitigating the progression of PD pathology.

Co-citation analysis identified two highly influential papers by Ronald B. Postuma’s team, published in Neurology in 2006 and 2009, demonstrate high betweenness centrality (Figure 7A). The 2006 paper, “Potential Early Markers of Parkinson Disease in Idiopathic REM Sleep Behavior Disorder” (0.10) (Postuma et al., 2006), was the first to propose RBD as an early marker of prodromal PD. The 2009 study, “Quantifying the Risk of Neurodegenerative Disease in Idiopathic REM Sleep Behavior Disorder” (0.13) (Postuma et al., 2009), expanded on these findings.

The Top 25 References with the Strongest Citation Bursts from 1993 to 2025 (Figure 7B) show that the earliest burst began in 2003, with the longest sustained burst occurring between 2009 and 2014. The reference with the strongest citation burst (83.66), still ongoing, is a 2019 review by Postuma’s team, published in Brain, titled “Risk and Predictors of Dementia and Parkinsonism in Idiopathic REM Sleep Behavior Disorder: A Multicentre Study” (Postuma et al., 2019). Currently, seven papers are experiencing citation bursts (Postuma et al., 2019; Galbiati et al., 2019; Hogl et al., 2018; Heinzel et al., 2019; Dauvilliers et al., 2018; Horsager et al., 2020; Miglis et al., 2021), they address phenoconversion risk, biomarkers, and prodromal disease trajectories. Cluster analysis of references (Figures 7C,D) indicates that cluster #3 (phenoconversion), cluster #5 (cognitive dysfunction), and cluster #6 (alpha-synuclein) are the most recent, reflecting evolving research interests and emerging trends in the field. The 2019–2025 Top 25 References with the Strongest Citation Bursts (Figures 7E,F) indicates that five papers are experiencing ongoing citation bursts (Cesari et al., 2022; Gilat et al., 2022; Dauvilliers et al., 2018; Arnaldi et al., 2021; Chahine et al., 2021), focusing primarily on RBD diagnosis, prediction of phenoconversion, and pharmacological treatment. These studies emphasize the current state of drug therapies, advocating for larger-scale clinical trials to evaluate long-term efficacy and stressing the significance of early identification and intervention in RBD management.

Given the database integration constraints, citation burst detection and reference clustering were based solely on the WOSCC data. Reference cluster analysis (Figures 7G,H) reveals that the latest clusters are #0 (prodromal), #5 (melatonin), #6 (graph theory), and #8 (risk disclosure). This analysis not only highlights current research hotspots but also traces the evolution of these topics over time, illustrating emerging trends. From the 1993–2025 Cluster Dependency Analysis (Figure 7C), two main trends emerge: (1) The first trend shows an evolution from cluster #0 (narcolepsy) to cluster #9 (non-motor symptoms), subsequently advancing to the latest clusters #3 (phenoconversion), #5 (cognitive dysfunction), and #6 (alpha-synuclein); (2) The second trend indicates that cluster #4 (Lewy body disease) developed into cluster #1 (mild cognitive impairment) and cluster #12 (neuropsychiatric symptoms), which further evolved into the latest clusters #3 (phenoconversion), #6 (alpha-synuclein), and #2 (glucocerebrosidase). In the past 5 years, cluster #2 (glucocerebrosidase) and cluster #5 (cognitive dysfunction) have advanced into cluster #5 (melatonin) (Figure 7G). Concurrently, cluster #3 (phenoconversion) has evolved into cluster #6 (graph theory), and cluster #6 (alpha-synuclein) has transformed into cluster #8 (risk disclosure), reflecting the dynamic shifts in research priorities and methodologies within the field.

Keyword analysis serves as a key bibliometric tool for tracking the evolution and identifying research trends in the field revealing current hotspots(Ellegaard and Wallin, 2015; Chen and Song, 2019). Category-based keyword analysis indicates that 12 categories exhibit high betweenness centrality (Figure 8A) with Neurosciences (0.82) Clinical Neurology (0.42) Engineering Biomedical (0.39) and Rehabilitation (0.24) playing key roles underscoring their central influence in RBD-related PD research. Co-occurrence analysis reveals that in addition to “Parkinson’s disease” and “RBD,” frequently occurring keywords predominantly focus on clinical manifestations pathophysiology early diagnosis and the link between RBD and cognitive impairment or neurodegenerative disease progression.

Keyword burst detection using VOSviewer was performed for keywords with more than five co-occurrences (Figure 8B). Recent keywords include fMRI MRI organization atrophytreatment outcome trial control symptoms predicts early transition mitochondrial dysfunction hypertension prodromal Parkinson disease skin biopsy clock genes and cerebrospinal fluid signaling emerging research hotspots in recent years.

CiteSpace cluster analysis of keywords from 1993 to 2025 (Figure 8C) shows that behavior disorder (0.12) and Lewy body disease (0.10) possess high betweenness centrality, indicating their key roles in PD and RBD research, aligning with the trends observed in reference analysis (Figures 7C,G). The analysis produced eight clusters, with the earliest being cluster #1 (dementia with Lewy bodies) and cluster #3 (restless legs syndrome). Cluster #1 initially centered on terms like RBD and degenerative disease, with prodromal dementia recently emerging. The latest cluster is cluster #4 (MRI), featuring new keywords such as organization, predictors, machine learning, cerebellum, and cells. The largest cluster, cluster #0 (non-motor symptoms), includes recent terms like critique and fluctuations. Active clusters include cluster #0 (non-motor symptoms), cluster #1 (dementia with Lewy bodies), cluster #2 (REM sleep without atonia), cluster #4 (MRI), and cluster #5 (Gaucher disease) (Figure 8D). Keyword clustering analysis for 2019–2025 (Figures 8E,F) identifies clusters such as cluster #0 (functional connectivity), cluster #5 (skin biopsy), and cluster #6 (structural MRI) as not only active but also rapidly expanding, indicating that these areas represent emerging and evolving research trends in the field.

Keyword bursts offer valuable insights into keyword trends helping detect shifts in research focus over time. Analysis of the top 25 keywords with the strongest citation bursts from 1993–2025 (Figure 8G) reveals that “levodopa” and “behavior disorder” were among the first to exhibit bursts while “parasomnia” (1996–2013) had the longest burst duration. The keyword “multiple system atrophy” showed the highest burst strength (18.6). Currently “biomarkers” and “gender differences” are still experiencing active bursts.

For the 2019–2025 period (Figure 8H), keywords such as “melatonin,” “Japanese version,” “epidemiology,” “community,” “reliability,” “index,” “normative data,” and “pramipexole” are undergoing active bursts. These terms reflect a focus on the diagnosis, epidemiology, treatment, and research methodologies related to RBD, suggesting that controlling RBD could potentially reduce or delay the onset of Parkinson’s disease, indicating their prominence as current research hotspots.

Tracing the evolution of research across different timeframes reveals a shift in focus. Early studies concentrated on the clinical features of RBD and its association with other neurodegenerative disorders, suggesting shared pathological mechanisms between RBD and these diseases. Later research shifted toward establishing RBD as a potential early marker for neurodegenerative conditions, particularly Parkinson’s disease and Lewy body dementia. The use of drugs like “levodopa” and dopamine agonists demonstrated that RBD is associated with dopamine transporter dysfunction. Additionally, international workshops on RBD facilitated global discussions, shedding light on phenomena such as hallucinations and the delayed onset of RBD, reinforcing its role as an early indicator of neurodegeneration. More recent studies have delved into the relationship between RBD and the non-motor symptoms of PD, as well as the role of biomarkers in tracking neurodegenerative disease progression. Non-motor symptoms such as olfactory dysfunction, daytime sleepiness, and mild cognitive impairment have gained prominence. Multicenter studies have expanded the understanding of RBD by uncovering aspects such as metabolic network activity and gender differences, further elucidating the relevance of biomarkers in predicting disease progression.

Collectively, the bibliometric analyses demonstrate a rapidly expanding and increasingly interdisciplinary research landscape in RBD-related Parkinson’s disease. Global collaboration has intensified, core research groups and institutions have emerged, and thematic evolution reveals a transition from descriptive clinical studies toward biomarker-driven prediction and early intervention research.

RBD is not only a prevalent non-motor symptom of PD but also a recognized early marker for its onset (Schutz et al., 2022; Byeon, 2020; Benbir-Senel et al., 2024; Berg et al., 2021; Borghammer and Van Den Berge, 2019; Postuma et al., 2019; Galbiati et al., 2019; Baumann-Vogel et al., 2020; Xie et al., 2021). With the global aging population expanding, there is growing attention on RBD for its role in the early prevention and effective intervention of PD. This bibliometric analysis provides a systematic overview of the intellectual landscape and evolutionary trajectory of research at the intersection of RBD and PD. This study maps how scientific attention has shifted, consolidated, and diversified across disciplines, methodologies, and translational priorities over time. The rapid increase in the number of publications characterized by a strong linear growth trend, reflects the expanding recognition of RBD as a critical prodromal manifestation of synucleinopathies rather than merely a secondary non-motor symptom. Importantly, this growth trend mirrors patterns reported in previous bibliometric analyses of prodromal PD and non-motor symptoms (Shi and Yih, 2024; Zhao et al., 2024; Sun et al., 2024; Pan et al., 2024; Cinpolat and Akar, 2023), which similarly identified RBD as a central research axis linking sleep medicine, neurology, and neurodegeneration. The expansion of the field appears to coincide with a broader paradigm shift in PD research, from motor-centric models toward early, system-level disease conceptualizations.

The geographical and institutional distribution of publications reveals a highly centralized research structure. Currently, countries like the USA, Japan, and England lead RBD-related PD research, with the USA and Japan being the earliest pioneers in the field, forming the core of international collaboration networks. This pattern likely reflects the long-standing investment in sleep medicine research in the USA and the demographic pressures of population aging in Japan, which have driven large-scale longitudinal cohort studies of RBD. In contrast, China has emerged as the fastest-growing and most promising contributor in recent years, likely driven by demographic aging, increased investment in neuroscience research, and expanding participation in international collaborations. However, network analyses indicate that cross-regional collaboration remains relatively fragmented, highlighting an underdeveloped global research integration despite the field’s translational ambitions.

Co-authorship and co-citation analyses further demonstrate influential researchers and institutions. Leading institutions in this field include McGill University, Mayo Clinic, University of Barcelona, and Philipps University Marburg, while prominent researchers such as Jean-Francois Gagnon, Ronald Postuma, Bradley F. Boeve, and Birgit Hoegl have made significant contributions. The prominence of these institutions and researchers reflect the centrality of long-term prospective cohort designs and phenoconversion studies in advancing RBD research.

Co-citation and reference clustering analyses further demonstrate that the knowledge base of RBD-related PD research is organized around a number of highly influential conceptual milestones. Seminal studies by Ronald B. Postuma’s team, published in Neurology in 2006 and 2009 (Postuma et al., 2006; Postuma et al., 2009), were among the first to suggest that RBD may serve as an early marker of prodromal PD. Current research primarily revolves around two key themes: narcolepsy and Lewy body disease. Recent studies experiencing citation bursts emphasize themes such as α-synuclein pathology, cognitive dysfunction, and ethical considerations of risk disclosure, underscoring the maturation of the field from descriptive observation toward anticipatory and prognostic frameworks (Postuma et al., 2019; Galbiati et al., 2019; Hogl et al., 2018; Heinzel et al., 2019; Dauvilliers et al., 2018; Horsager et al., 2020; Miglis et al., 2021). Cluster analysis of references has identified several emerging research clusters, further underscores the methodological demands of prodromal PD research relies heavily on sustained follow-up, standardized assessments, and interdisciplinary expertise. From a bibliometric standpoint, these works function as intellectual anchors, shaping subsequent research on prodromal diagnosis, phenoconversion prediction, and translational trial design.

Bibliometrics aims to uncover research hotspots and development trends, providing valuable insights for scholars (Ellegaard and Wallin, 2015; Chen and Song, 2019). Keyword co-occurrence, burst detection, and timeline analyses collectively reveal a clear transition in research focus over time. Current research is primarily focused on Neurosciences, Clinical Neurology, Biomedical Engineering, Rehabilitation, and Experimental Medicine. Early research concentrated on the clinical features of RBD and its associations with neurodegenerative diseases, suggesting an exploratory phase focused on shared pathological mechanisms. Later studies shifted toward investigating RBD as an early marker for neurodegenerative disorders like PD and Lewy body disease. Recent research emphasizes the link between RBD and PD’s non-motor symptoms, effective interventions, and the role of biomarkers in neurodegenerative progression, highlighting the significance of early detection and intervention in slowing PD’s pathology. This thematic shift aligns RBD research with broader trends in neurodegenerative disease research, in which early-stage identification and risk modeling have become central objectives. The field has increasingly converged on a translational core that emphasizes objective markers capable of linking sleep-related manifestations to underlying synuclein pathology. The bibliometric evidence thus suggests not only topic diversification but also methodological maturation, with growing reliance on longitudinal designs, multimodal assessment strategies, and hypothesis-driven investigations.

While this study offers valuable insights into the research hotspots and trends in Parkinson’s disease-related RBD through bibliometric analysis, several limitations should be acknowledged. First, although both the WoSCC and Scopus were used to enhance coverage and reduce database-specific bias, publications indexed exclusively in other databases may still be underrepresented. Second, newly published articles require time to accumulate citations, and the focus on literature from the past 5 years may underrepresent emerging research areas or the contributions of early-career scholars. Ongoing efforts to monitor developments in this field are necessary. Third, only English-language publications were included, which may introduce language bias and underestimate contributions from non-English-speaking regions, especially in a rapidly expanding global research field. Lastly, bibliometric visualizations inevitably reflect the analytical frameworks adopted. While established software and standard configurations were applied, the findings should be interpreted as knowledge mappings rather than definitive representations of the field. Advances in bibliometric software could help mitigate this limitation in future studies.