Global research hotspots and trends in exercise interventions for rheumatoid arthritis over the past two decades: A bibliometric and visualization study

Abstract

Rheumatoid arthritis (RA) is a prolonged multifactorial autoimmune disease of unknown etiology. With the global population aging, the incidence of RA is increasing, highlighting the need for more effective treatments. Exercise interventions have been recognized as safe and effective for managing pain, improving function, and reducing fatigue in RA patients. However, the existing literature in this field lacks a thorough, organized, and clear line of analysis. In this study, we conducted a comprehensive analysis of the 20-year literature on exercise interventions for RA, aiming to identify hotspots and cutting-edge trends. Our objective is to provide subsequent researchers with valuable ideas and references. Using Cite Space, VOS viewer, and R-bibliometrix software for visualization and analysis, we compiled the main dataset from the web of science database, consisting of 1790 articles on exercise interventions in RA published between 2000 and 2023. Among these articles, the United States contributed the highest number of papers (433), while Karolinska Institutet ranked first institutionally with 90 papers. The study focused on the keyword’s quality of life, cardiovascular disease, aerobic exercise, social support, psychology, and multidisciplinary care. The research highlighted the importance of clinical efficacy studies that investigate different types of exercise modalities (cardiorespiratory aerobic, resistance, aquatic, and neurological) either alone or in combination, to improve pain and function and reduce cardiovascular disease risk in patients with RA. Additionally, sedentary behavior, fatigue, and multidisciplinary care were identified as potential areas for further research. Overall, this study provides a scientific perspective on exercise interventions for RA and offers valuable insights for academics, funding organizations, and policymakers.

1. Introduction

Rheumatoid arthritis (RA) is a chronic, systemic, autoimmune, inflammatory illness and is the most frequent type of inflammatory arthritis.^[1,2] The global prevalence of RA is approximately 0.5% to 1%,^[3] with women having a 3 times higher prevalence compared to men.^[4] The typical age of onset falls between 40 and 60 years.^[5] The main symptoms include joint pain, swelling, morning stiffness, and functional limitation. As the disease progresses, joint damage becomes progressive and irreversible, potentially leading to disability.^[6] It is concerning that almost 1 to 3rd of RA patients lose their jobs,^[7,8] and the rate of work disability is 10 times higher than the general population.^[9] In 2005, the estimated overall cost of RA in the US was $19.2 billion.^[10] In addition, patients with RA have a 50% to 60% increased risk of premature death from cardiovascular disease.^[11] The adverse effects of cardiovascular disease on RA are believed to be similar in magnitude to those of diabetes.^[12] Therefore, it is crucial to not only treat the symptoms of RA but also implement a comprehensive strategy that includes cardiovascular risk management. Identifying beneficial, sustainable, and cost-effective interventions is important for improving the quality of life for people with RA and reducing the burden on individuals and society. Physical activity refers to any skeletal muscle-produced movement that consumes energy.^[13] The World Health Organization recommends that the general population engage in 150 minutes of moderately intense aerobic activity or 75 minutes of vigorous aerobic exercise per week.^[14] Physical inactivity contributes to 6% of the global total death rate, making it the 4th most significant risk factor.^[15] Meanwhile, regular physical activity is recommended as a component of cardiovascular risk management in the general population. Over the past 2 decades, significant progress has been made in clinical research on various types of exercise to improve pain, function, fatigue, and quality of life in RA patients. With the emergence of new theories and technologies, new opportunities and challenges for RA treatment have arisen. However, there is currently no bibliometric study in this field that investigates the research hotspots and trends related to exercise intervention in RA. Therefore, the aim of this study is to analyze and track the hotspots, frontiers, and evolutionary paths in order to gain a rapid and in-depth understanding of the field. The findings of this study can provide valuable references and support for researchers in this area.

Cite Space and VOS viewer are powerful tools for generating co-citation networks based on reference citations, allowing researchers to explore the structure of specific research areas. These tools support visual exploration and knowledge discovery in bibliographic databases. A progressive knowledge domain visualization approach enables the visual mapping of highly cited and critical literature.^[16] Cite Space is particularly effective in identifying research hotspots and trends within the academic field, while VOS viewer excels in analyzing researcher collaboration and co-authorship networks. In our study, we utilized VOS viewer 1.6.18 and Cite Space 6.1.R6 for bibliometric analyses. Additionally, we employed R-Studio-based R-bibliometrix 4.6.1 (https://posit.co/), Pajek 5.16, and Scimago Graphica 1.0.26 for multimodal geovisualization.

2. Methods

2.1. Search strategy and data retrieval

Bibliometric research is a quantitative method utilized for analyzing and evaluating scientific literature. One of the widely used academic databases for bibliometric analysis is Web of Science, which is highly regarded for its comprehensiveness and reliability.^[17] We exported the data collected by Web of Science in “plain text” format, which consisted of complete records and references. Each file contained 500 records and was named “download”. These files were then imported into the input folder of citespace software for visualization analysis. The analysis involved selecting node types such as country (region), institution, author, journal, cited author, keyword, and cited literature for co-occurrence or cluster analysis. Finally, a visualization map was generated.

Table S1, Supplemental Digital Content, http://links.lww.com/MD/K660 provides detailed information on the search strategy. The literature selected for this study adhered to specific inclusion and exclusion criteria. These criteria included: Publication dates between 2000-01-01 and 2023-02-10; Literature types limited to Article and Review; Literature written in English; No restrictions on the species or organisms studied; and Removal of duplicates to ensure a unique dataset. Ethical approval was not required as the articles did not contain any personal information about the patients. The related workflow diagram, as shown in Figure 1.

Figure 1.

Workflow diagram.

2.2. Literature selection

The literature was independently assessed by 2 reviewers. Initially, the title and abstract of the article were screened, followed by the selection of inclusion or exclusion based on predetermined criteria. In cases where there was disagreement, a third assessor reviewed the entire paper and made the final decision on inclusion. This third assessor had the authority to veto any decision.

2.3. Bibliometric analysis

Our bibliometric analysis study utilized several tools, including Cite Space version 6.1.R6 (https://citespace.podia.com/), VOS viewer version 1.6.18 (https://www.vosviewer.com/), R-Studio based R-bibliometrix version 4.6.1, Pajek version 5.16 (http:// mrvar.fdv.uni-lj.si/pajek/), and Scimago Graphica version 1.0.26 (https://www.graphica.app/).

Cite Space version 6.1.R6 (developed by Prof Chen C at Drexel University, Philadelphia, PA) is a software tool specifically designed for identifying research hotspots and trends within the academic field. It achieves this by analyzing citation relationships and evolutionary trends of the literature, allowing for the identification of key research themes and widely cited papers. In our study, Cite Space was used to conduct a visual analysis covering various aspects such as country and institutional distribution, journal citations, reference analysis, and keyword and citation bursts. The specific parameters used were as follows: a time span of 2000 to 2023, a time scale of 1-year, a filtering criterion of “Top N,” and a threshold of 50. Additionally, the “Pathfinder” and “Pruning sliced networks” methods were chosen as the clipping connection methods to simplify the network structure and highlight important features. For more details, please refer to Table S2, Supplemental Digital Content, http://links.lww.com/MD/K661.

VOS viewer version 1.6.18 (Leiden University, Leiden, the Netherlands) is a visualization software developed by Prof Van Eck and Prof Waltman. It focuses on analyzing the co-occurrence of words in literature to discover common themes and concepts. Through co-occurrence frequency analysis, researchers can construct a co-occurrence network diagram for visualization. The software also examines the co-occurrence frequency and temporal information of keywords in the literature, revealing the evolution process of research topics and the development trend of keywords. This helps researchers understand the dynamic nature of the academic field.^[18] In this study, we utilized VOS viewer to analyze and visualize the distribution of countries, institutions, authors, and co-cited authors. We also explored keyword co-occurrence and coverage networks. The specific parameters used were as follows: the keyword inclusion criterion required a minimum of 5 occurrences, the association strength method was chosen for normalization, the clustering resolution and minimum cluster size were both set to 1, and small clusters were merged. Additionally, the minimum strength of clustered links was set to 0, and the maximum number of links was limited to 1000. For more information, please refer to Table S3, Supplemental Digital Content, http://links.lww.com/MD/K662.

R-Studio-based R-bibliometrix version 4.6.1, Pajek version 5.16, and Scimago Graphica version 1.0.26 are powerful tools for multimodal and multidimensional geovisualization. This study utilized these tools to emphasize the inter-cooperative network relationships among various countries or regions.

3. Results

3.1. Characteristics and trends of the volume of articles issued

A total of 1841 literature related to exercise intervention RA studies were retrieved by searching the core database of Web of Science. From these, 31 documents published outside the timeframe of January 1, 2001 to February 10, 2023 were excluded. Additionally, 11 documents that were not categorized as Article or Review were excluded. Finally, 9 documents that were not in English were also excluded. As a result, a total of 1790 documents were included, as shown in Figure 1 of the flow chart. The annual publishing volume in different nations is depicted in Figure 2A. From 2000 to 2023, there has been a consistent rise in the number of research papers on exercise interventions for RA. Notably, there was a significant surge in publications in 2015 and 2021, with a 19% and 21% increase compared to the previous year, respectively. The field reached its peak in 2021 with a total of 156 publications in that year. In terms of overall publications, the Americas have the highest percentage, followed by developed Western countries like the United Kingdom and the Netherlands. We conducted a polynomial fit analysis to determine the associations between the year of publishing and the overall number of articles. The coefficients of determination (R2) for total papers, articles, reviews, and randomized controlled trials were 0.9491, 0.9553, 0.8482, and 0.7462, respectively. Based on our research, we projected that approximately 180 papers would be published in 2025, consisting of approximately 138 articles, 42 reviews, and 19 randomized controlled trials (RCTs), as shown in Figure 2B. Overall, the field of rheumatology and rehabilitation medicine has experienced significant growth, leading to more comprehensive research. However, it is evident that despite the increase in publications each year, there is still a lack of high-quality RCT trials.

Figure 2.

(A) Bibliometric analysis of the WoS core database output. (B) Trends in publications in the field of exercise intervention RA research and the corresponding polynomial fit curves. (C) Map of the intensity of collaboration across countries/regions in the field of exercise intervention RA research. Networks of national/regional cooperation in this area. Circles represent countries and the size of the circle indicates the number of publications. Different colors represent different clusters and connecting lines represent international collaboration between countries. The thickness of the connecting lines indicates the strength of the collaboration. Link: https://tinyurl.com/2cshht4f. (D) Populated graph of the collaborative bibliometric analysis of countries in the field of exercise intervention RA research. The darker the color, means more posts, and the connecting lines represent partnerships. (E) Mechanism co-occurrence network diagram for exercise intervention RA. The circle in the chart indicates the volume of articles issued, the larger the circle indicates the more articles issued by the institution, the thickness of the purple outer circle represents the centrality of the institution, and the connecting line represents the existence of cooperation or co-occurrence. (F) Map of the intensity of collaboration across institutions in the RA research area of exercise intervention. Networks of institutions cooperation in this area. Circles represent institutions and the size of the circle indicates the number of publications. Different colors represent different clusters, while the lines connecting the projects represent international collaboration between institutions. The thickness of the connecting lines indicates the strength of the collaboration. Link: https://tinyurl.com/2b5zw3yk. RA = rheumatoid arthritis.

3.2. Country or regional cooperation characteristics

A total of 79 nations or areas were included in the visualization, with 30 having more than 20 articles. Table 1 provides information on the top 10 nations with the highest quantity of papers. These top 10 countries are all developed countries, highlighting a notable disparity in research output between advanced and developing nations. The United States stands out as the country with the most publications (433), indicating its significant influence in this discipline. Figure 2C showcases international cooperation among 48 countries/regions that have collaborated on at least 5 publications. The darker color in the filled graph represents stronger cooperation with other countries Figure 2D. Notably, the figure reveals robust research collaborations between the UK, the USA, and the Netherlands, while collaborations between Europe and North America are considerably stronger compared to those with Australia and Asia.

Table 1.

Top 10 high-impact countries and institutions for exercise intervention RA research.

Country	Number of articles issued	Institution	Number of articles issued
USA	433	Karolinska Institutet	90
UK	264	University of Copenhagen	60
Netherlands	189	Harvard University	54
Sweden	168	Karolinska University Hospital	54
Canada	137	University of London	51
Denmark	90	Leiden University	50
Germany	86	UDICE-French Research Universities	48
Italy	84	Leiden University - Excl LUMC	46
France	80	Assistance Publique Hopitaux Paris	42
Australia	80	Leiden University Medical Center (LUMC)	42

RA = rheumatoid arthritis.

3.3. Characteristics of research institution cooperation

Research on exercise therapies in RA was conducted by a total of 931 different institutes. Among these institutes, 37 published 20 or more articles on the subject. Table 1 provides information on the top 10 organizations based on the number of articles published, while Figure 2E visually represents this data. The top 10 institutions, which have the highest number of publications, contribute to 30% of the total number of publications. Additionally, Figure 2F displays the international collaboration between 88 institutions that have collaborated on at least 10 publications. Notably, 3 institutions, namely Karolinska Institutet, University of Birmingham, and University of Copenhagen, have established strong research collaborations with some of the other institutions.

3.4. High-impact author collaboration features

The visualization atlas included a total of 4717 writers, with 20 authors having 10 or more publications. Table 2 displays information on the top 10 scholars based on the number of articles and the top 10 scholars based on citation frequency. The figure indicates that there is collaboration among the author teams, especially among the highly productive authors. However, there is a lack of authors with high centrality in the literature, as shown in Figure 3A and B. The 3 most prominent researchers in the field of exercise interventions for RA, namely Kitas GD, Metsios GS, and Opava CH, each had 27 publications and were highly active and influential writers. According to the author co-citation analysis conducted by VOSviewer, the top 3 authors in terms of co-citation frequency for exercise intervention RA studies were Wolfe F (404 times), Hakkinen A (370 times), and Metsios GS (344 times). Figure 3C illustrates the network of co-cited author relationships, which includes 142 authors with a citation frequency of 50 or more. A comprehensive analysis of published papers and coauthors determined that the team represented by Kitas GD, Stavropoulos-Kalinoglou A, and Metsios GS from the University of Birmingham, UK, focused on examining the clinical efficacy of different types of exercise training to improve cardiorespiratory fitness and reduce the risk of cardiovascular disease in patients with RA.^[19–21] Molecular biology of exercise to reduce the risk of cardiovascular illness in RA patients.^[7,22–24] Another team, led by Opava CH, Nordgren B, and Demmelmaier I at Karolinska University Hospital, Sweden, focused on investigating the clinical efficacy of physical activity in maintaining and improving long-term health in RA patients.^[25,26] Additionally, a study explored the role of healthcare workers in effectively guiding RA patients to increase their physical activity.^[27,28] Lastly, Lamb SE, Williams MA, and Williamson E at the University of Warwick, UK, conducted research on the therapeutic effectiveness of hand exercise training in RA patients.^[29,30]

Table 2.

Top 10 high-impact authors of exercise intervention RA studies.

Author	Country	Institution	Number of articles issued	Author	Frequency of citations	Country
Kitas GD	UK	University of Birmingham	27	Arnett FC	259	USA
Metsios GS	UK	University of Wolverhampton	27	Wolfe F	220	USA
Opava CH	Sweden	Karolinska University Hospital	27	Hakkinen A	188	Finland
Mannerkorpi K	Sweden	University of Gothenburg	16	Metsios GS	174	UK
Duda JL	UK	University of Birmingham	16	Sokka T	172	Finland
Nordgren B	Sweden	Karolinska Institutet	16	Fries JF	142	USA
Demmelmaier I	Sweden	Karolinska Institutet	15	Aletaha D	141	Austrian
van Zanten JJCSV	Sweden	University of Birmingham	15	De Jong Z	132	Netherlands
Gossec L	France	University of Sorbonne	13	Prevoo MLL	117	Netherlands
Hakkinen A	Finland	Jyväskylä Central Hospital	13	Smolen JS	109	Austrian

RA = rheumatoid arthritis.

Figure 3.

(A) Author co-occurrence network diagram of the exercise intervention RA. Circles in the plot indicate the number of publications, with larger circles indicating more publications by the author, and connecting lines indicating the existence of a collaborative or co-occurring relationship, link: https://tinyurl.com/24cgdm6y. (B) Plot of high-yielding authors of exercise intervention RA over time. The top twenty most prolific researchers in the field and their publications. The larger the node, the more literature published. The darker the color, the more citations. The color represents the number of publications and the color represents the number of citations per year. (C) Cluster visualization of author co-citation analysis generated based on VOSviewer software. Each node represents an author, and the size of each circle is determined by the author’s co-citation, http://tinyurl.com/23pml9ax. RA = rheumatoid arthritis.

3.5. High-impact publications and citation patterns

The rankings of the number of publications and the H-index are displayed in Table 3 and Figure 4A respectively. The top 10 journals account for 26.14% of the total publications. Among these, only 2 journals with an impact factor >10 are Annals of The Rheumatic Diseases (27.973) and Arthritis and Rheumatism-Arthritis Care and Research (15.483). These 2 journals contribute to 17.3% of all papers published in the top 10 overall. Therefore, it is challenging to publish relevant papers in high-impact factor journals. According to the journal co-citation analysis conducted using VOSviewer, the top 3 co-cited journals for exercise intervention in RA research are Annals of The Rheumatic Diseases (5206 times), Journal of Rheumatology (4960 times), and Rheumatology (2878 times). Figure 4B presents the network of co-cited journal relationships, including 153 journals with a citation frequency of 100 or more. The biplot overlay’s-colored trails connecting journal groupings highlight the connection between citing and cited journals in terms of citations, providing insights into the citation trajectory and information flow.^[31] In the field of Medicine/Medical/Clinical, articles often reference journals from the fields of Sports/Rehabilitation/Sport, Health/Nursing/Medical, and Molecular/Biology/Genetics. Figure 5 provides more details on the typical cited and cited journals within each cluster. For example, the Journal of Rheumatology, Annals of The Rheumatic Diseases, Rheumatology, and Arthritis Care and Research are commonly referenced periodicals in the field of Sports/Rehabilitation/Sport.

Table 3.

Status of high-impact journals for exercise intervention RA research, 2000-2023.

Journal name	Total number of articles	Total number of applications	The average number of citations	IF (2022)	JCR (2022)	H-index
Arthritis Care and Research	81	560	6.91	5.178	Q2	33
Rheumatology	68	583	8.57	7.046	Q1	35
Journal of Rheumatology	56	285	5.09	5.346	Q2	28
Rheumatology International	54	210	3.89	3.58	Q4	16
Arthritis and Rheumatism-Arthritis Care and Research	49	752	15.35	15.483	Q1	33
Clinical Rheumatology	46	169	3.67	3.65	Q3	17
Annals of The Rheumatic Diseases	32	368	11.50	27.973	Q1	29
Disability and Rehabilitation	29	70	2.41	2.439	Q1	13
Scandinavian Journal of Rheumatology	27	145	5.37	3.057	Q3	13
Bmc Musculoskeletal Disorders	26	113	4.35	2.562	Q2	14

IF = impact factor, RA = rheumatoid arthritis.

Figure 4.

(A) H-index of high-impact journals for exercise intervention RA research, 2000-2023. (B) Cluster visualization of journal co-citation analysis generated based on VOSviewer software. Each node represents a journal, and the size of each circle is determined by the journal’s co-citation, link: https://tinyurl.com/2nuk55qt. RA = rheumatoid arthritis.

Figure 5.

Biplot overlay of cited and cited journals in the field of exercise intervention RA research. Cited journals are on the left, cited journals are on the right, and the line represents the citation status. RA = rheumatoid arthritis.

3.6. Keyword visualization analysis

3.6.1. Hotspot analysis of keyword co-occurrence clustering.

The investigation of research trends and the identification of research hotspots heavily rely on keyword analysis.^[18] Figure 6A and B, as well as Table 4, display the terms with the highest co-occurrence. Terms with high centrality and frequency indicate the research hotspots over time. Figure 7 illustrates the clustering network of keyword co-occurrences in this area. Using the typical LLR algorithm, a total of 18 groups were formed, and the keyword clustering analysis revealed that greater aggregation led to more homogeneous relationships between studies.^[32] The cluster number and size have an inverse connection, with cluster number 0 representing the largest cluster and so on. The clusters are intricately intertwined and closely linked. The current research hotspots in the field, identified through keyword co-occurrence and cluster analysis, include quality of life, cardiovascular disease, classification, low back pain, randomized controlled trials, aerobic exercise, social support, and psychology.

Figure 6.

(A) Keyword co-occurrence map of exercise intervention RA. The circles in the figure represent keywords, and the larger the circle indicates the higher the frequency of keywords. The dark to light colors represent the years from far to near, the connecting lines represent the links between the keywords, and the thickness of the purple outer circle represents the centrality of the keywords. (B) VOSviewer-based clustering view of exercise intervention RA keywords, link: https://tinyurl.com/2jlzb7zc. RA = rheumatoid arthritis.

Table 4.

Campaign intervention RA high frequency keywords and centrality TOP10.

Keywords	Frequency	Keywords	Centrality
Rheumatoid arthritis	1339	disease activity	0.09
Exercise	729	diseases	0.08
Quality of life	276	disability	0.07
Disease activity	215	association	0.07
Disease	159	aerobic exercise	0.07
Health	158	program	0.06
Pain	136	mortality	0.06
Randomized controlled trial	121	bone mineral density	0.06
Adults	115	responses	0.06
Women	114	risk	0.05

RA = rheumatoid arthritis.

Figure 7.

Keyword co-occurrence clustering map of exercise intervention RA. The dark to light colors represent the years from far to near, and the connecting lines represent the links between keywords. RA = rheumatoid arthritis.

3.6.2. Analysis of research trends in emergent word emergence.

The analysis revealed several emerging words that are frequently used in the field of exercise intervention RA, as shown in Figure 8A. The most prominent mutation occurred in the term “muscle strength” (13.01), followed by “long term” (10.63), and “bone mineral density” (10.23). Notably, the keywords that experienced the highest level of mutation in the past 5 years include “fatigue” (2019–2023), “classification” (2019–2023), “criteria” (2019–2023), “depression” (2018–2021), and “adults” (2020–2021). By utilizing Burst emergence analysis and examining the dynamic temporal evolution of keywords, we can gain a comprehensive understanding of current and future research trends in the field of exercise intervention RA. The emergent words, such as meta-analysis, fatigue, classification, and criteria, have remained significant and are likely to continue being hot topics.

Figure 8.

(A) Keyword emergent graph of exercise intervention RA. Note: The figure of “▂” is a 1-year keyword mark, “▂” is the emergence of the year of the word. (B) Co-cited literature on exercise interventions for RA. The circles in the figure represent cited literature (showing the first author’s name), and the larger the circle, the more citations the literature has received. The thicker the circle is, the more citations the literature has received. The line represents co-citations among the literature and the thickness of the purple outer circle represents the centrality of the literature. (C) Clustering of co-cited literature for exercise intervention RA. The dark to light colors represent the years from far to near, and the connecting lines represent the links between keywords. (D) Emergence of key literature on exercise interventions for RA. The “▂” in the figure are the years in which the unexpected citations appeared, and the “▂” in the figure are the nodes in which the number of citations for the unexpected citations suddenly rose, arranged in chronological order from top to bottom. RA = rheumatoid arthritis.

3.7. Visual analysis of key documents

3.7.1. Analysis of research hotspots in key literature.

A total of 1790 documents were found, which contained a total of 5837 citations in this category. Figure 8B, Tables 5 and 6 display the rankings of the top 10 highly cited and highly centralized documents. To identify the research hotspots and evolutionary trajectories in the field of sports intervention RA, it is essential to analyze the literature with the highest citation frequency and co-cite important nodes. The studies can be divided into 3 groups based on their type: clinical trials, clinical guidelines, and reviews. The studies with centrality ranking 1st, 3rd, 4th, 8th, 9th, and 10th, as well as citation frequency ranking 4th, 8th, 9th, and 10th, are classified as clinical studies. On the other hand, the studies with citation frequency ranking 1st and 3rd are classified as clinical guideline studies. Lastly, the studies with centrality ranking 2nd, 6th, and 7th, as well as citation frequency ranking 2nd, 5th, 6th, and 7th, are categorized as review studies. The research techniques employed in these studies include the double-blind, randomized controlled method (used in studies with centrality ranking 1st, 3rd, 8th, and 9th, as well as citation frequency ranking 8th and 9th) and the experimental method (used in studies with centrality ranking 4th and 10th, as well as citation frequency ranking 10th). The assessment of treatment outcomes mostly relied on visual analog rating scales, such as aerobic capacity (VO2max), disease activity score (28), functional score (The MOS item short from health survey-36), health assessment questionnaire, and pain visual analogue scale (VAS). These indicators were used to evaluate changes in the values of the visual analogue rating scales after a predetermined follow-up period, in order to demonstrate the therapeutic effects of various forms of exercise on RA. All clinical trials findings consistently show that regular physical activity has a positive impact on the long-term prognosis of RA through exercise interventions of varying durations. The values of the VAS before and after therapy also showed significant differences. However, in the 9th study mentioned in the literature,^[41] researchers conducted a moderate-intensity exercise intervention in a warm water pool and observed no significant changes in aerobic capacity and the MOS item short from health survey-36 fitness components. However, they did find significant improvements in muscular endurance and flexibility after a 3-month follow-up.

Table 5.

Top 10 ranking of cited literature centrality of exercise intervention RA.

Author	Centrality	Year	Title	Periodicals
van den Berg MH[33]	0.12	2006	Using internet technology to deliver a home-based physical activity intervention for patients with rheumatoid arthritis: A randomized controlled trial	Arthrit Rheum-Arthr
Steultjens EMJ[34]	0.11	2002	Occupational therapy for rheumatoid arthritis: a systematic review	Arthrit Rheum-Arthr
de Jong Z[35]	0.04	2003	Is a long-term high-intensity exercise program effective and safe in patients with rheumatoid arthritis? Results of a randomized controlled trial	Arthritis Rheum
Baillet A[36]	0.04	2009	A dynamic exercise programme to improve patients’ disability in rheumatoid arthritis: a prospective randomized controlled trial	Rheumatology
Baillet A[37]	0.03	2010	Efficacy of cardiorespiratory aerobic exercise in rheumatoid arthritis: meta-analysis of randomized controlled trials	Arthrit Care Res
Eurenius E[38]	0.03	2005	Physical activity, physical fitness, and general health perception among individuals with rheumatoid arthritis	Arthrit Rheum-Arthr
Lee J[39]	0.03	2012	Public health impact of risk factors for physical inactivity in adults with rheumatoid arthritis	Arthrit Care Res
Brodin N[40]	0.03	2008	Coaching patients with early rheumatoid arthritis to healthy physical activity: a multicenter, randomized, controlled study	Arthrit Rheum-Arthr
Bilberg A[41]	0.03	2005	Moderately intensive exercise in a temperate pool for patients with rheumatoid arthritis: a randomized controlled study	Rheumatology
Knittle KP[42]	0.03	2011	Effect of self-efficacy and physical activity goal achievement on arthritis pain and quality of life in patients with rheumatoid arthritis	Arthrit Care Res

RA = rheumatoid arthritis.

Table 6.

Top 10 ranked frequency of cited literature for exercise intervention RA.

Author	Frequency of citations	Year	Title	Periodicals
Osthoff AKR[43]	75	2018	2018 EULAR recommendations for physical activity in people with inflammatory arthritis and osteoarthritis	Ann Rheum Dis
van Zanten JJCSV[44]	40	2015	Perceived Barriers, Facilitators and Benefits for Regular Physical Activity and Exercise in Patients with Rheumatoid Arthritis: A Review of the Literature	Sports Med
Agca R[45]	38	2017	EULAR recommendations for cardiovascular disease risk management in patients with rheumatoid arthritis and other forms of inflammatory joint disorders: 2015/2016 update	Ann Rheum Dis
Stavropoulos-Kalinoglou A[19]	30	2013	Individualised aerobic and resistance exercise training improves cardiorespiratory fitness and reduces cardiovascular risk in patients with rheumatoid arthritis	Ann Rheum Dis
Tierney M[5]	28	2012	Physical activity in rheumatoid arthritis: a systematic review	J Phys Act Health
Metsios GS[7]	28	2018	Physical activity, exercise and rheumatoid arthritis: Effectiveness, mechanisms and implementation	Best Pract Res Cl Rh
Verhoeven F[15]	26	2016	Physical activity in patients with rheumatoid arthritis	Joint Bone Spine
Thomsen T[46]	24	2017	The efficacy of motivational counseling and SMS reminders on daily sitting time in patients with rheumatoid arthritis: a randomized controlled trial	Ann Rheum Dis
de Jong Z[35]	21	2003	Is a long-term high-intensity exercise program effective and safe in patients with rheumatoid arthritis? Results of a randomized controlled trial	Arthritis Rheum
Demmelmaier I[25]	20	2013	Current and maintained health-enhancing physical activity in rheumatoid arthritis: a cross-sectional study	Arthrit Care Res

RA = rheumatoid arthritis.

3.7.2. Analysis of research hotspots for co-occurrence clustering of key literature.

Based on co-cited literature, cluster analysis can be used to identify subfields that represent important areas of research.^[47] The classic LLR technique was employed to generate a total of 13 clusters. The clustering of the literature reveals the sub-thematic directions that are currently being extensively researched in the field. The cluster number and size have an inverse relationship, with #0 representing the largest cluster and so on see Figure 8C. The lighter-colored clusters in the field indicate research trends related to muscle strength, multidisciplinary care, control evaluation, and joint protection.

3.7.3. Analysis of research trends emerging from key literature.

Burst citations are significant works that have been frequently referenced over time, highlighting important topics and trends see Figure 8D. According to the Web of Science database, the EULAR guidelines for exercise in individuals with osteoarthritis and inflammatory arthritis, which were released in 2018 by Osthoff AKR, had the highest burst intensity among the 25 most co-cited literature. These guidelines emphasize the importance of reaching a consensus on relevant research and education agendas, as well as promoting physical exercise as an essential aspect of conventional care for these diseases (41.72).^[43]Another noteworthy publication is a narrative review by van Zanten JJCSV in 2015, which discusses the barriers, benefits, and facilitators of physical activity and training in patients with RA (22.41).^[44] Lastly, Agca R’s 2017 EULAR guidelines for managing cardiovascular disease risk in patients with RA and other inflammatory joint diseases are also among the top burst citations (20.11).^[45] The literature on exercise interventions in RA can be categorized into 4 main sections based on their chronological order. The first section (2000–2008) focuses on studies examining the effectiveness and safety of long-term high-intensity exercise interventions in RA,^[35,48] as well as the increased risk of cardiovascular disease in RA patients due to lack of physical activity.^[22] The 2nd section (2008–2013) discusses the clinical efficacy of cardiorespiratory aerobic exercise interventions in RA,^[49] with a meta-analysis-based reevaluation of various exercise interventions.^[37,50] The 3rd section (2013–2018) includes studies on the clinical efficacy of individualized aerobic combined with resistance exercise interventions for RA,^[19,51] as well as research on fatigue-related risk factors and intervention treatments in RA.^[50,52,53] The 4th section focuses on research guidelines in the field, particularly highlighting the low physical activity rates among patients with RA. It emphasizes the need for healthcare professionals to be more aware of the crucial role of exercise interventions in RA. The section also provides recommendations for rational exercise prescription and suggests research and educational agendas.^[43,45] Notably, studies have been conducted to explore the use of Internet-based information devices in reducing sedentary patient time and lowering the risk of cardiovascular disease^[46,54] (2018-present). The intensity of the literature burst in parts III and IV has continued until now. After carefully examining the highly cited articles, it is possible to predict future study trends. These trends include evaluating the long-term efficacy of different levels and types of physical activity, assessing the effectiveness of reducing sedentary behavior over time, identifying optimal strategies to increase patient compliance with physical activity interventions (including motivation and beliefs), determining response and nonresponse markers to exercise interventions, identifying disease contraindications in various exercise domains (such as cardiovascular, strength, neuromotor, and flexibility), and exploring new technologies like Internet-based tools for controlling and monitoring changes in sedentary behaviors.

3.7.4. Research hotspots and trend analysis of thematic maps.

The thematic maps generated based on the R-Bibliometrix package are displayed as a 2-dimensional matrix see Figure 9. By analyzing the quadrant in which the critical bubbles are located, it can be inferred that the hot spots in this field include quality of life, health, pain, disability, expression, oxidative stress, interleukin-6, and activation. It is possible that themes such as c-reactive protein, necrosis-factor protein, necrosis-factor-alpha, tumor-necrosis-factor, and body-mass index may undergo future developments or become extinct. In the context of RA, further in-depth studies will be required on topics such as exercise, physical activity, disease activity, muscle strength, inflammation, cardiovascular disease, and skeletal muscle.

Figure 9.

Thematic map of exercise interventions for RA. RA = rheumatoid arthritis.

4. Discussion

4.1. Research hotspots in exercise interventions for rheumatoid arthritis

This study explored the research hotspots through keyword co-occurrence clustering, key literature co-occurrence clustering, and thematic map hotspots. These techniques address various interrelated topics within a specific discipline.^[55] The main keywords identified as research hotspots in this field include quality of life, cardiovascular disease, aerobic exercise, social support, psychology, multidisciplinary care, and muscle strength.

Although pharmacological therapies have had a substantial impact, physical therapy and exercise programs continue to be crucial parts of managing RA.^[56,57] In fact, a study found that a significant number of adults with RA (42%) are inactive.^[39] However, a meta-analysis of studies has shown that cardiopulmonary aerobic exercise can improve function, quality of life, and pain in RA patients.^[36] It is worth noting that there is inconsistency in standardizing the intensity of aerobic exercise for middle-aged and older patients with RA.^[58–60] Considering that cardiovascular events greatly affect the outcomes of RA, incorporating aerobic exercise as a supplementary treatment is of utmost importance in managing the condition.^[61,62] Research has demonstrated that aerobic exercise not only enhances cardiovascular health and quality of life for patients but also reduces disability and pain associated with RA.^[63] Resistance training has shown promise in sedentary patients, particularly in those with RA. A recent meta-analysis involving 512 RA patients found that resistance training significantly reduced disease activity score-28 scores, erythrocyte sedimentation rate scores, and 50-foot walk times.^[64] Currently, most clinical studies in this area combine aerobic exercise with resistance exercise to treat RA patients, and these studies have shown positive effects in terms of aerobic capacity, endurance, strength, and pain relief.^[19,51,65] RA is associated with accelerated atherosclerosis, and endothelial dysfunction is an initial cardiovascular abnormality before the onset of atherosclerosis.^[66] Physical exercise plays a role in cardiovascular risk and the development of atherosclerosis. Several potential mediators of the impacts of physical activity include increased blood flow, better antioxidant defenses, and higher endothelial nitric oxide synthase activity.^[24] Recent research has shown that a well-designed program of aerobic exercise and strength training 3 times per week can lead to improvements in endothelial function.^[67,68] Physical exercise has also been associated with better cardiac autonomic function, specifically in terms of the heart. A personally designed training program consisting of 2 or 3 45-minute sessions per week over a 3-month period has been found to improve heart rate variability.^[69] Furthermore, long-term intensive exercise programs have been shown to improve functional capacity in patients with RA without causing additional damage to large joints, unless there is already considerable baseline damage.^[34] Therefore, exercise intervention in aquaaerobic or exercise in warm water has been identified as a beneficial option for patients with RA.^[70,71] Several studies have shown that moderate-intensity exercise in water can significantly improve muscle endurance and flexibility in RA patients after a 3-month follow-up period.^[41] However, it is worth noting that there is a lack of research on flexibility exercise and neuromotor interventions for RA in this field. Only 1 RCTs has been found, which examined the positive effects of a neuromotor training program on neuromotor performance in RA patients.^[72] In addition to aquaaerobic exercise, other intervention techniques such as yoga, tai chi, and dance classes have also shown potential benefits for RA patients. For instance, after 6 weeks of intervention, some studies have reported significant improvements in self-esteem, VAS, and health assessment questionnaire scores with 2 yoga lessons per week.^[73,74] Similarly, tai chi^[75] and dance classes^[76,77] have been suggested as potentially helpful interventions.

4.2. Global trends in exercise interventions for rheumatoid arthritis

Based on an analysis of keyword emergence, key literature emergence, and thematic map trends in the field, the identified keywords in research trends for this study were sedentary behavior, fatigue, multidisciplinary care, standards, classification, meta-analysis, health-related quality of life, joint protection, disease activity, and cardiovascular disease. The analysis of the most recent keyword outbreak revealed 3 emerging trends in research on exercise interventions for RA.

Sedentary behavior: Sedentary behavior is defined as low energy expenditure activities performed while sitting or lying down, with a metabolic equivalent of ≤ 1.5.^[78] In RA patients, the risk of premature death from cardiovascular disease is increased by 50% to 60%.^[11] Research has shown that RA patients spend between 71% and 92% of their waking hours in a seated position,^[79] whereas healthy individuals dedicate only 62% of their waking time to sedentary activities.^[80] Despite the recommendation of incorporating exercise alongside medication, RA patients often fail to engage in the recommended amounts of moderate-to-high-intensity resistance and aerobic exercise.^[81] Therefore, it may be more practical to encourage the substitution of inactive habits with light-intensity activities in RA patients, rather than solely focusing on increasing moderate-to intense physical activity,^[82,83] which can include activities of daily living such as standing, walking, and housework. Studies have shown that taking scheduled breaks from sedentary activities can lead to improvements in plasma glucose, insulin levels, and resting blood pressure.^[84] In addition, Internet-based monitoring of RA patients has shown that reducing sedentary behavior has positive effects on health-related quality of life, pain, physical function, self-efficacy, and cholesterol levels in the short and long-term (4, 18 months).^[46,85] However, improving patients compliance with physical activity is crucial. Studies have found that a lack of motivation and related beliefs are strongly associated with inactivity in 65% of RA patients.^[39] On the other hand, support from healthcare workers, friends, and family has been identified as a positive factor in promoting physical activity among RA patients.^[44] Therefore, further research should be conducted to explore strategies for incorporating physical activity/exercise to reduce sedentary behavior in patients with RA.

Fatigue: Fatigue is a commonly reported effect among RA patients, It is described as feeling tired during or after usual activities, or feeling insufficient energy to start these activities.^[86] However, this description fails to capture the sensation reported by RA patients that weariness is unpredictable and not related to energy consumption. Unlike normal fatigue,^[87] RA fatigue is often seen as debilitating and limits normal functioning.^[88,89] Patients place a high priority on minimizing fatigue because it significantly impacts their quality of life.^[90] RA fatigue is influenced by various factors, including those related to the disease (pain, disability, joint damage), personal factors (work/caregiving obligations, health, loss of social support, environment), and cognitive and behavioral factors (anxiety, stress, sickness beliefs, despair).^[91] Reports suggest that between 41% and 80% of RA patients experience fatigue.^[92,93] However, fatigue in RA is currently not well managed. The underlying mechanisms of RA fatigue are not known. One potential nonpharmacologic treatment for fatigue in adults with RA is exercise. Systematic reviews suggest that several forms of exercise are beneficial in the treatment of weariness in RA.^[52,94] However, there is only 1 RCTs that uses fatigue as a primary endpoint for aerobic exercise in RA, and that trial found that an exercise intervention significantly reduced fatigue.^[95] Additionally, a recent meta-analysis by Rongen-van Dartel, which focused on land-based aerobic exercise in individuals with RA, found that the degree of fatigue improvement after ≤ 12 weeks of exercise was statistically significant, but the improvement was not statistically significant after ≥ 24 weeks.^[50] Therefore, more high-quality RCTs exploring the effectiveness of different physical activity/exercise strategies to improve fatigue in patients with RA are needed in the future.

Multidisciplinary care: The principle of interventions in RA is to recommend exercise therapy in multidisciplinary guidelines, particularly for patients with early RA. Patients and healthcare professionals should receive instruction on exercise programs for RA, along with reminders of the benefits of physical activity and exercise.^[96,97] However, there are currently no defined criteria for the most appropriate exercise prescription for patients with RA who have different ages, aerobic capacity, and obesity rates.^[98,99] Recent research indicates that nearly all participating rheumatologists, clinical nurse specialists, and physical therapists consider it essential to encourage patients with RA to engage in regular physical exercise.^[100] However, some healthcare providers lack confidence in prescribing exercise and have limited knowledge of appropriate referral options for RA,^[100–102] except for sports medicine physicians and physical therapists.^[103] Physiotherapists, who specialize in exercise, possess extensive knowledge of exercise physiology and the ability to adapt exercise prescriptions to different conditions. In addition, there is insufficient consideration is given to patient-preferred models of public health service delivery, such as supervised/unsupervised, face-to-face/online, individual/group, and booster strategies.^[43] is given to patient-preferred models of public health service delivery, such as supervised/unsupervised, face-to-face/online, individual/group, and booster strategies.^[104] Therefore, it is crucial to involve professionals from various disciplines. Furthermore, it is important to provide unified health professional education on specific exercise activity programs, health communication, and referral processes.

4.3. Limitations of the study

This research utilized only English-language literature articles from the Web of Science database’s core dataset. It should be noted that the current visualization software can only analyze a single database, which imposes limitations on literature retrieval. Consequently, high-quality publications from other databases or in other languages on the topic may have been overlooked. It is important to acknowledge that the generation of visualization maps lacks a standardized setting process for time partitioning, screening criteria (threshold), and cropping method, which may introduce bias.

5. Conclusion

This study provides a comprehensive analysis of exercise intervention research in the field of RA over the past 20 years. Using bibliometric and visual analysis, the study examines various aspects including the influential nations, organizations, publications, and authors. The findings reveal that USA, Karolinska Institutet, Arthritis Care & Research, and Metsios GS have had significant influence in this area. The research highlights the clinical efficacy of different types of exercise modalities (such as cardiorespiratory aerobic, resistance exercise, aquatic exercise, and neurological exercise) in improving pain, function, and reducing the risk of cardiovascular illness in RA patients. Additionally, sedentary behavior, fatigue, and multidisciplinary care have emerged as potential new areas of research.

Acknowledgments

The authors thank all the authors of the original studies included in this analysis.

Author contributions

Conceptualization: Meng Chen.

Data curation: Jie Xu, Meng Chen, Yingli Yu.

Formal analysis: Meng Chen, Yingli Yu.

Funding acquisition: Yingli Yu, Liugang Tang, Yuandong Cheng.

Investigation: Jie Xu, Xiaobing Luo.

Methodology: Jie Xu, Liugang Tang.

Project administration: Yingli Yu.

Resources: Liugang Tang, Xiaobing Luo.

Software: Jie Xu, Meng Chen.

Supervision: Xiaobing Luo, Yuandong Cheng.

Validation: Liugang Tang, Yuandong Cheng.

Visualization: Jie Xu.

Writing – original draft: Jie Xu, Meng Chen.

Writing – review & editing: Jie Xu, Meng Chen.