Bibliometric and visualization analysis of global research trends on immunosenescence (1970–2021)

BACKGROUND
Immunosenescence, the aging of the immune system, leads to a decline in the body's adaptability to the environment and plays an important role in various diseases. Immunosenescence has been widely studied in recent years. However, to date, no relevant bibliometric analyses have been conducted. This study aimed to analyze the foundation and frontiers of immunosenescence research through bibliometric analysis.


METHODS
Articles and reviews on immunosenescence from 1970 to 2021 were obtained from the Web of Science Core Collection. Countries, institutions, authors, journals, references, and keywords were analyzed and visualized using VOSviewer and CiteSpace. The R language and Microsoft Excel 365 were used for statistical analyses.


RESULTS
In total, 3763 publications were included in the study. The global literature on immunosenescence research has increased from 1970 to 2021. The United States was the most productive country with 1409 papers and the highest H-index. Italy had the highest average number of citations per article (58.50). Among the top 10 institutions, 50 % were in the United States. The University of California was the most productive institution, with 159 articles. Kroemer G, Franceschi C, Goronzy JJ, Solana R, and Fulop T were among the top 10 most productive and co-cited authors. Experimental Gerontology (n = 170) published the most papers on immunosenescence. The analysis of keywords found that current research focuses on "inflammaging", "gut microbiota", "cellular senescence", and "COVID-19".


CONCLUSIONS
Immunosenescence research has increased over the years, and future cooperation and interaction between countries and institutions must be expanded. The connection between inflammaging, gut microbiota, age-related diseases, and immunosenescence is a current research priority. Individualized treatment of immunosenescence, reducing its negative effects, and promoting healthy longevity will become an emerging research direction.


Introduction
Immunosenescence is the slow reduction in immune system function caused by aging, influencing the composition, quantity, and function of immune organs, immune cells, and cytokines (Pawelec, 2018). Immunosenescence is a complicated and multifaceted process defined by progressive changes in innate and adaptive immune system processes that result in the development of various diseases (Rodriguez et al., 2022). In terms of innate immunity, aging is frequently associated with compromised anatomical barriers, decreased neutrophil chemotaxis and phagocytosis, diminished clearance of apoptotic corpses by macrophages, and diminished antigen-presenting potential of dendritic cells (Rodriguez et al., 2022;Toutfaire et al., 2017;Man et al., 2015). The adaptive immune system is integrally linked to and tuned by the innate immune system. Consequently, the innate immune system changes with age, much like the adaptive immune system. The main features of immunosenescence include continuous low-grade inflammation, loss of ability to respond to antigens, an increase in the prevalence of autoimmunity, and a loss of memory responses (Goronzy and Weyand, 2013;Adamczyk-Sowa et al., 2022). In addition, multiple immune cell types undergo phenotypic alterations. Furthermore, latent and chronic viral infections, such as human cytomegalovirus, can affect the immune system and lead to immunosenescence with age (Lee et al., 2022).
Immunosenescence is related to increased morbidity and death among the elderly (Martínez de Toda et al., 2021). Although aging should not be regarded as a disease, it is a key risk factor for developing chronic age-related diseases such as cardiovascular, metabolic, and neurodegenerative diseases-all related to low-grade inflammation (HAP et al., 2019;Furman et al., 2019). In the elderly, these illnesses can ultimately result in organ failure and death. As immunosenescence progresses, elderly people become increasingly susceptible to infectious illnesses and cancer. The elderly population is at a greater risk of contracting influenza and COVID-19 as well as dying from them. Notably, individuals with chronic (inflammatory) diseases have an elevated risk of developing severe COVID-19 and dying from the virus . Consequently, there is a connection between immunosenescence and aging-related illnesses. Immunosenescence research can provide feasible methods for preventing and treating chronic diseases and improving health in older adults.
Currently, bibliometrics is a popular tool for systematically reviewing any field of study. Bibliometrics not only examines qualitatively and statistically the contribution and collaboration of authors, institutions, countries, and journals but also the evolution and development trends in science investigation Ke et al., 2020). There has been an increasing number of studies on immunosenescence over the past few decades. However, to our knowledge, there has been no systematic assessment of existing research on this topic. Hence, this study aimed to systematically summarize and visually assess publications on immunosenescence based on the Web of Science and using CiteSpace (Synnestvedt et al., 2005) and VOSviewer (van Eck and Waltman, 2010) software to comprehend the frontiers and emerging trends of research.

Data source and search strategy
This was bibliometric research that did not include clinical studies or patient consent. Therefore, neither an ethics committee nor an institutional review board was required. The Web of Science Core Collection (WoSCC) bibliographic database is among the world's largest and most complete electronic databases for scientific publications. Data were obtained from the Science Citation Index Expanded (SCI-E) of the WoSCC of Clarivate Analytics on October 13, 2022. All searches were conducted on the same day to reduce potential bias caused by database upgrades. We performed pertinent pretests and improved the retrieval technique to ensure the integrity and precision of the results. The search formula was set to TS = ("immunosenescence" OR "immun* senescence" OR "immun* aging" OR "aging immun*" OR "immun* ageing" OR "ageing immun*"). Immunosenescence-related terms entered into the WoS engine were retrieved from the Medical Subject Headings (MeSH) in PubMed, and the wildcard character "*" was substituted for any number of characters to conduct the most exhaustive search of related studies. The following were the criteria for selection: (Pawelec, 2018) The literature search was conducted between January 1, 1970, andDecember 31, 2021;(Rodriguez et al., 2022) the only allowed document types were "article" and "review"; (Toutfaire et al., 2017) the language type was set to only English. Documents listed below were excluded: (Pawelec, 2018) If the literature type was a meeting abstract, editorial material, early access, correction, letter, or preceding paper; (Rodriguez et al., 2022) if the literature was not related to immunosenescence; (Toutfaire et al., 2017) if they were unpublished documents lacking adequate information. The workflow of this study is illustrated in Fig. 1.
A total of 3763 papers were exported, including 2735 "articles" and 1028 "reviews". Two authors (YT and CZ) extracted the data independently, including the annual study numbers, countries, institutions, authors, journals, citations, and keywords. Disagreements between the two reviewers were resolved by discussion with a third reviewer. The search results were then saved as "Plain Text" with the information from "Full Record and Cited References".

Data processing and analysis
This descriptive study reported the variables as numbers and percentages. No comparisons were made; hence, no P-values were calculated. Bibliometric analysis was performed using the R language (version 3.6.3), bibliometrix R package (version 3.1.4), CiteSpace (version 6.1.R3), VOSviewer (version 1.6.18), and Microsoft Excel 365. The H-index identifies a researcher who has authored at least H articles that have been referenced at least H times-created to measure the influence of scientific research . Bibliometric (https: //bibliometric.com/) is an open-source instrument for quantitative research in scientometrics. We chose the "bibliometrix" R package to map the three-field plot and show the links between the main elements.
CiteSpace is a bibliometric and visual analysis software program that identifies collaboration, important spots, internal structures, possible trends, and evolutions in a given scientific subject (Chen, 2004). We utilized CiteSpace to evaluate the dual map of journals, reference timeline, citation bursts, keyword timeline, and keyword bursts. In CiteSpace visualization, the size of each node is proportional to the cooccurrence frequency, and the links represent co-occurrence associations. The colors of the nodes and lines indicate the number of years. CiteSpace was configured with the following settings: time slicing from January 1970 to December 2021 and years per slice = 3. Throughout the analysis of keyword co-occurrence and bursts, minimum spanning trees and pruning sliced networks were chosen. The pathfinder and pruned sliced networks were chosen to examine the co-cited reference cluster and other indicators. The software utilized cluster analysis to determine correlations, relying on nominal terms to identify study hotspots, enabling researchers to locate mutation words on the map, academic hotspots, and comprehend the direction of the study. There are three Clustering algorithms: latent semantic indexing (LSI), log-likelihood ratio (LLR), and mutual information (MI).
Based on bibliographic data, we selected VOSviewer software to identify the co-occurrence of countries/regions and institutions, prolific journals, co-cited journals, keywords, authors, co-cited authors, and knowledge graphs. VOSviewer focuses on the graphical depiction of bibliometrics and is especially beneficial for displaying big bibliometrics in an understandable manner (van Eck and Waltman, 2010). In the cluster map, nodes of the same color belong to the same cluster. In addition, the link displays a co-occurrence relationship, and the thickness of the connection is related to the amount of papers co-authored by two scholars or the quantity of papers in which two keywords appear together (Zhang et al., 2022a).
Microsoft Office Excel 365 was used to construct the database and analyze the published papers. In addition, the impact factor (IF) and journal citation report (JCR) division of journals and the H-index of academics were collected from the Web of Science on October 13, 2022.

The trend of publication outputs
The temporal distribution of immunosenescence-related publications is shown in Fig. 2A. The total number of annual papers has increased from one in 1970 to 412 in 2021; most research was published in 2021 (412, 10.95 %). The growth of literary output was exponential, and the equation for the exponential curve was y = 1.3129e 0.1639x . The equation can predict the annual number of publications. The simulated curve closely resembled the annual publication growth trend with a strong coefficient of determination (R 2 = 0.9609), suggesting that the number of annual publications will keep increasing over the next several years. Before 2009, annual publication output increased slowly; however, since 2010, it has displayed a significant upward trend. Overall, these results show that research on immunosenescence has become a focal point and entered a period of rapid development.

Distribution of countries/regions
A total of 3763 articles were published in 102 different countries. The top 10 countries are listed in Table 1. The United States (1409, 37.44 %) contributed the most articles. This was followed by Italy (442, 11.75 %) and England (393, 10.44 %). The total number of citations for publications in the United States was 68,458-much higher than in other countries. This indicated that American publications of excellent quality had a certain reference value. Among the top 10 countries, Italy had the highest average number of citations per article at 58.50. Scholars from the United States had the highest H-index, suggesting that their articles were highly influential.  During the past two years, the number of publications in Italy, England, Germany, and China has increased significantly compared to previous years.
The cooperative relationships between the countries are shown in Fig. 3A. The United States had the greatest active influence, followed by Italy, Germany, England, and Spain. England and the United States enjoyed the closest cooperation, while the United States established cooperative connections with a majority of the world's countries. With a minimum number of documents of 20 as the screening criteria, a total of 32 countries were screened, and VOSviewer was used to map the citation relationships between countries (Fig. 3B). The density map reflected the quantity of articles in each nation (Fig. 3C). Word size, circle size, and red opacity are correlated with higher number of documents. The United States, Italy, Germany, and England are the major contributors to immunosenescence. Fig. 3D shows the map of citation relationships between countries over time. Publications in the United States and Italy were concentrated around 2013, while those in China, India, and South Korea were concentrated around 2017 (Fig. 3D).

Distribution of institutions
Concerning publication ranking, the top 10 contributive institutions are listed in Table 2. Among the top 10 institutions, 50 % were in the United States. The University of California System was the most productive institution with 159 (4.23 %) articles, followed by the University of Tübingen with 130 (3.45 %) articles, and Udice French Research Universities with 113 (3.00 %) articles. Thus, these institutions hold a prominent role in the field of immunosenescence research. The University of Bologna from Italy had the highest total citations (13,182) and average citations per article (146.47), indicating that the institution's publications were well regarded. The cooperative connection between institutions is shown in Fig. 4A, and each institution's collaboration time chart is shown in Fig. 4B. We screened 173 organizations with a minimum of 10 publications per organization. There were 173 nodes and three clusters on the network map, with the University of California System at the center of the nodes (Fig. 4A). Previous collaborations are shown by purple or blue nodes, whereas current collaborations are represented by yellow nodes (Fig. 4B). There was a strong collaborative relationship between some research institutions, such as the University of Tübingen, University of Sherbrooke, University of Birmingham, and University of Córdoba (Fig. 4). In contrast, the University of Western Australia and the Cranfield University had minimal cooperation with other institutions.

Authors and co-cited authors
A total of 15,835 authors were involved in immunosenescencerelated publications. The top 10 productive authors are listed in Table 3, contributing 570 publications and accounting for 15.15 % of all the papers. Pawelec G from Germany authored the most immunosenescence-related articles (n = 104), followed by Franceschi C (n = 66) and Goronzy JJ (n = 62). The collaborative network between authors is depicted in Fig. 5A, where the authors had published no fewer than ten papers. Co-cited authors were two or more authors who were simultaneously cited by one or more papers. Six authors were cited >700 times (Table 3). For the co-citation analysis, authors with at least 100 citations were grouped into four main clusters; Franceschi C was the most co-cited, followed by Pawelec G and Fulop T (Table 3 and Fig. 5B).

Journals and co-cited journals
The top 10 journals and co-cited journals that published articles on immunosenescence are listed in Table 4. Among them, Experimental Gerontology published the most articles (n = 170), followed by Frontiers in Immunology (n = 144) and Mechanisms of Ageing and Development (n = 114). The journal with the highest IF among the top 10 journals was Aging Cell (IF = 11.005). In addition, of the top 10 journals, four were in the Q1 JCR division, and five were in Q2 JCR division. We selected 59 journals on the basis of a minimum of ten research papers and plotted the journal network (Fig. 6A). The size of the node is proportional to the number of published articles, and the thickness of the link is proportional to the strength of the collaboration. The influence of journals is determined by their co-citation frequency, indicating a journal's significance in a certain study field. Six co-cited journals were cited >4000 times, including Journal of Immunology (11,987), Mechanisms of Ageing and Development (6099), Experimental Gerontology (4961), etc. The cocitation network was mapped by excluding journals with fewer than 500 co-citations (Fig. 6B). The size of the node is proportional to cocitation frequency, the Journal of Immunology has the most co-citations.
The dual-map overlay of journals revealed a relationship distribution among the academic journals (Fig. 7). The citing journals were located on the left, while the cited journals were located on the right, and the colored pathways represented the citation relationships. As shown, the orange path indicates that studies published in Molecular/Biology/Genetics journals were frequently cited in studies published in Molecular/ Biology/Immunology journals. The green path indicates that studies published in Molecular/Biology/Genetics journals tended to be cited primarily in Medicine/Medical/Clinical.

Co-cited references and references burst
Co-citation indicates that two or more articles are cited simultaneously by one or more papers. In general, the more frequently a piece of literature is quoted, the more significant it is in its subject. Supplementary Table 1 displays the top 20 most cited references retrieved, with each having been cited at least 107 times. The most co-cited reference (443 times) was a review published in the Annals of the New York Academy of Sciences by Franceschi et al. (2000), entitled "Inflammaging. An evolutionary perspective on immunosenescence". Fig. 8A shows a map of reference co-citations with the corresponding clusters. The top 19 co-citation clusters are presented in Supplementary  Table 2 and Fig. 8B. Fig. 8B describes 19 color blocks, each indicating a cluster of articles on the same subject. The areas with different colors represent the time when co-citation linkages first occurred in specific areas. By default, clusters with fewer than ten articles were filtered out and not displayed. For example, cluster #17 did not display. Filtering can make the clustering more representative and facilitate the identification of research hotspots and trends. The smaller the number, the larger the cluster. Cluster #0, cd95, was the largest cluster. The cluster analysis found three main research aspects: 1) intracellular signaling molecules, such as CD95, CD8, NKG2C, and signal transduction; 2) the relationship between viruses and immunosenescence, such as COVID-19, cytomegalovirus, herpes zoster, and HIV; 3) diseases related to immunosenescence, such as inflammaging, aging, caloric restriction, and rheumatoid arthritis. The timeline view of the references illustrates the progression of research hotspots over time (Fig. 8C). Each horizontal line represents a cluster. Each cluster may be labeled with the title terms, keywords, and abstract terms of papers citing that cluster. Cluster labels are assigned to the terms that occur most frequently in each cluster. The clusters indicated with purple represent appeared earlier, while the clusters marked with yellow color show current research concerns.
The label "References with citation bursts" denotes that the relevant material was often mentioned during a specific time frame. Fig. 8D shows the top 20 references with the strongest citation bursts. This figure also indirectly indicates the steady evolution of immunosenescence research, as each year has references with citation bursts, some persisting through 2021. This suggests that immunosenescence-related research may continue to expand in the future. The paper with the strongest burstness (strength = 53.32) was entitled "Immunosenescence and inflammaging as two sides of the same coin: friends or foes?" (Fulop et al., 2017), published in Frontiers in Immunology by Fulop T et al. in 2018, with citation burstness from 2018 to 2021.

Analysis of keywords and hotspots
Keywords are the core part of a research. By evaluating the keywords, we were able to identify the study subjects in a certain field and explore the research hotspots and trends. We used VOSviewer to create a network map of keywords, where the minimum number of keyword occurrences was 10 ( Fig. 9A). Based on the link strengths of item cooccurrences, the network was separated into three clusters. Cluster 1 (red) was the most significant, and the prominent keywords were immunosenescence, aging, inflammation, oxidative stress, and apoptosis, emphasizing age-related phenotypes. Cluster 2 (green) was primarily related to the molecular basis of immunosenescence, such as T cells, B cells, autoimmunity, thymus, telomere, CD57, and CD28. Cluster 3 (blue) focused on immunosenescence-related diseases, including the elderly, vaccination, influenza, COVID-19, asthma, and infection. The density visualization map of the keywords (Fig. 9B) revealed the keywords with high occurrence frequency were "aging", "immunosenescence", "inflammation", "T cell", "cytomegalovirus", and "cytokine". Fig. 9C depicts the overlay visualization map, which summarizes the keyword occurrences from a time-zone perspective. The keywords were "cytokines", "lymphocyte", "apoptosis", "signal transduction", "CD8", and "CD28" between 2008 and 2012. These keywords focus primarily on fundamental studies of immune aging. The keywords were "immunosenescence", "aging", "elderly", "T cell", "oxidative stress", "vaccine", and "cancer" between 2012 and 2016. After 2016, the keywords were mainly "inflammaging", "COVID-19", "SARS-CoV-2", "immunotherapy", "gut microbiota", "multiple sclerosis", "biomarkers", and "SASP" indicating a focus on aging-related phenotypes and age-related diseases. In the lower-right quadrant, the yellow nodes reflect the most recent average-appearing year, which may be the focus of future immunosenescence research. Fig. 9D depicts a three-field graph that connects the authors keywords and journals. Through this three-field graph it was possible to view the connections between the principal elements and their relationship was displayed immediately based on the strength of the connecting links (Yang et al., 2022). The authors Pawelec G Franceschi C and Larbi A were significantly linked to the keywords "immunosenescence" and "aging". Furthermore the experimental gerontology and mechanisms of aging and development connected most of the authors who related to the keywords "immunosenescence", "aging", and "T cells".
We conducted a keyword cluster analysis and obtained a cluster timeline view (Fig. 10A). Seventeen clusters were formed (Supplementary Table 3), reflecting the changes in hot research directions over time. "Gene expression", "disease", "infection", and "age" were large clusters with multiple articles. The top 20 keywords with the highest burst strengths are shown in Fig. 10B. We noticed that the keyword with the highest number of citation outbreaks was "lymphocyte" (strength = 23.68), followed by "old mice" (20.09) and "signal transduction" (13.09). The keywords "cellular senescence", "health", and "gut microbiota" had the most recent outbreak citations (2018-2021).

General information
Bibliometric analysis has been widely applied to evaluate trends and advancements in numerous academic domains. Over the past few decades, annual publications on immunosenescence research have continuously increased (Fig. 2). From 1970 to 2021, a total of 3763 articles related to immunosenescence were published in 1084 academic journals by 15,835 writers from 3812 institutions in 102 countries/regions, according to the WoSCC database.
In the country/region analysis, the numbers of publications and citations were two crucial variables. According to Table 1, the United States dominated the global study of immunosenescence with the largest number of relevant publications (average of 48.59 citations per article), while the University of California System published the greatest number of studies on immunosenescence (Table 2). In their most recent studies, they demonstrated that age-related deterioration of T cell immunity was associated with sex-dimorphic elevation of N-glycan branching, identifying several possible therapeutic targets in T cell aging (Mkhikian et al., 2022). There was limited cooperation between some institutions, which could be damaging to the growth of academic research over the long term. Consequently, we propose that research institutes should engage in broad cooperation and communication to collaboratively advance immunosenescence development.
From the perspective of the authors and co-cited authors (Table 3), Pawelec G not only produced the most immunosenescence-related articles, but he was also among the top three co-cited authors, demonstrating his excellent contribution to immunosenescence research. He recently published a review explaining the association between myeloid-derived suppressor cells (MDSCs) and human aging, arguing that age-related expansion of MDSCs may be an outcome of inflammaging (Pawelec et al., 2021). Franceschi C had the highest number of citations and co-citations, publishing a vital paper in 2018 that was constructive and profoundly significant for aging and inflammaging research (Franceschi et al., 2018). For the top journals, Experimental Gerontology, Frontiers in Immunology, Mechanisms of Ageing and Development, and the Journal of Immunology can be considered the core journals for immunosenescence publication (Table 4).
The co-citation relationship of publications changes with time, with document co-citation networks predicting the development and evolution of a discipline (Qin et al., 2022;Wang et al., 2019a). Among the top 20 co-cited articles, most types of literature were reviews, focusing on the concept, mechanism, and characteristics of immunosenescence, as well as the relationship between immune aging, inflammation, and agerelated diseases (Franceschi et al., 2000;Trzonkowski et al., 2003;Franceschi and Campisi, 2014;Gruver et al., 2007;Franceschi et al., 2007). Among them, the paper with the greatest number of co-cited references was Franceschi C's "Inflamm-aging. An evolutionary perspective on immunosenescence" published in 2008. In this article, the authors defined inflam-aging, which is a major feature of the aging  process, as a decline in the ability to deal with diverse stressors and a progressive increase in a proinflammatory state (Franceschi et al., 2000). Later, Franceschi C et al. also published the second most co-cited paper (Franceschi et al., 2007). They proposed that inflammaging and anti-inflammaging are significant factors in global aging and longevity (Franceschi et al., 2007). In addition, an article by Fulop T had the strongest citation burst. They concluded that immunosenescence can be considered adaptive or remodeling rather than merely destructive (Fulop et al., 2017). Thus, through co-occurrence analysis, cluster analysis, and timeline view analysis of the literature, we may determine that the most recent literature citations concentrate on COVID-19, cellular senescence, inflammaging, viruses, and the elderly (Fig. 8).

The hotspots and trends
Keywords can help us rapidly identify the distribution and development of immunosenescence research hotspots (Zhang et al., 2022b). From Fig. 9C (co-occurrence analysis of keywords) and Fig. 10B (keywords with citation bursts), it can be seen that immunosenescence research has shifted from basic to clinical research. We can conclude that inflammaging, gut microbiota, and age-related diseases are current research hotspots

The role of aging-related phenotypes in immunosenescence
Numerous age-related phenotypes, such as senescence-associated secretory phenotype (SASP), inflammaging, shortened telomeres, and decreased telomerase activity-risk factors for age-related disorders-are related to immunosenescence.

SASP.
The SASP secreted from senescent adaptive immune cells is a proinflammatory phenotype consisting of inflammatory factors, chemokines, growth factors, and extracellular matrix proteases that continually accumulate as the number of senescent cells in diverse organs increases (Zhu et al., 2015;Guan et al., 2020). The SASP is a ubiquitous feature of cell senescence. As shown in Fig. 9C, the keyword "SASP" has been extensively studied recently, and the role of SASP in immunosenescence is gradually being valued. The SASP has been demonstrated to affect immune system functions, including senescent cell clearance (Ovadya et al., 2018). In addition, SASP promotes cellular senescence in a paracrine manner, promoting senescence in surrounding cells that have no aging characteristics (Hubackova et al., 2012;Hoare and Narita, 2013;Acosta et al., 2013). SASP causes inflammation, recruits immune cells, and affects both cells and tissues (Prata et al., 2018). The central signaling pathway for SASP formation may also be shared by several types of senescent cells, converging on the transcription factor NF-κB-a significant regulator of inflammation with a crucial role in the initiation of SASP (Chien et al., 2011;Meyer et al., 2017;Salminen et al., 2012). Notably, SASP is also associated with changes in the blood levels of certain metabolites, including tryptophan and steroids, whose involvement in immunity is well-described; thus, immunosenescence and SASP are intimately linked to other age-related illnesses (Rodriguez et al., 2022).

Inflammaging.
Inflammaging refers to sterile, persistent, lowgrade, chronic inflammation that increases with age (Franceschi et al., 2000;Ray and Yung, 2018). Inflammaging is one of the seven pillars of the aging process and a hallmark of the most prevalent age-related disorders (Kennedy et al., 2014). With age, the immune system becomes less susceptible to infections and the efficiency of vaccines typically declines (Zhang et al., 2016). The innate immune system is characterized by a quick yet relatively non-specific response to infection or injury. The levels of nucleic acids, cardiolipin, mitochondria, and heat shock proteins from cell death or damage increase with age, triggering innate immune receptors to release proinflammatory cytokines (Teissier et al., 2022;Biagi et al., 2010). Viral infection (Kohli et al., 2021;Cox and Lord, 2021;Camell et al., 2021) and bacterial infection (AEM et al., 2019) are other causes of senescence, resulting in protracted and damaging inflammation. In addition, aging-related dysbiosis of the gut microbiota is a driving force for homeostasis of the immune system and a significant source of inflammatory stimuli (Mangiola et al., 2018;Biagi et al., 2016). Compared with younger people, the elderly are characterized by higher production of proinflammatory cytokines such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α (Fagiolo et al., 1993). Together with IL-8 and c-reactive protein, these cytokines are the most significant circulating indicators of inflammation (Ferrucci and Fabbri, 2018) and have been identified as an immune signature for age-related chronic inflammation (Sayed et al., 2021). Many of the top 20 co-cited publications were related to inflammaging (Supplementary Table 3 The top 10 authors and co-cited authors that contributed publications on immunosenescence.    (Blackburn et al., 2015). Telomere length and telomerase activity are reduced with aging in lymphocytes. A shorter telomere may cause DNA damage and cell cycle arrest, ultimately leading to impaired cell activity and ineffective elimination of pathogens . The immune response is harmed by the downregulation of telomerase activity, which also stimulates old cells during cloning (Zhou et al., 2021).

Immune cells and aging
Immunosenescence affects not only the innate immune system--comprising natural killer (NK) cells and macrophages-but also the acquired immune system, which consists of T and B cells. Hence, senescence of the immune system reflects senescence of diverse immune cell subpopulations (Lian et al., 2020). The network map of keywords (Fig. 9A) shows that immune cells, particularly T and B cells, play a crucial role in the mechanism of immunosenescence.

T cells. T cells, the most important cells in acquired immunity,
were one of the most frequently occurring keywords (Fig. 9B), showing their relevance in immunosenescence research. T cells mature in the thymus and move to peripheral lymphoid organs to proliferate and differentiate into memory and effector T cells in response to antigenic stimulation, exerting a significant influence on immune system function (Zhu et al., 2010). T cells experience senescence due to the lack of the costimulatory molecules CD27 and CD28, reduced growth factor IL-2, and increased production of proinflammatory cytokines Pawelec, 2014). Senescent T cells can trigger inflammatory reactions by contacting other immune cells, secreting proinflammatory cytokines, or directly affecting target tissue, resulting in tissue injury and contributing to the pathophysiology of aging (Sharabi and Tsokos, 2020). Furthermore, senescent T cells diminish the ability to identify novel pathogens and the vaccination response, thereby increasing susceptibility to the infectious process (Yang et al., 2020).

B cells.
B cells are the only cells that produce antibodies and play a special role in immunity (Wang et al., 2019b). Aging influences the subset distribution of mature B cells and impairs their activation in response to stimulation (Pinti et al., 2016;Bulati et al., 2017). With aging, the percentage and quantity of CD19+ B cells in the peripheral blood decline, and B cell activity is compromised (Cepeda et al., 2018). This decreases the body's ability to respond to antibodies and manufacture high-affinity antibodies, thereby increasing the risk of infectious diseases, cancer, and autoimmune diseases (Rodrigues et al., 2021).

Other cells.
Macrophages are strong immunoregulatory innate immune cells that play an essential role in immunological defense and inflammation regulation (Weiskopf and Weissman, 2015). Macrophages also exhibit age-related functional changes. Aged macrophages exhibit impaired phagocytosis, altered sensitivity to lipopolysaccharide, and diminished capacity to move to areas of infection (van Beek et al., 2019).
NK cells are a subset of lymphocytes that contribute to innate immunity, and distinct alterations in NK cell subsets with advancing age are often accompanied by diminished functional activity (Naumova et al., 2016). The main feature is the low responsiveness to cytokines, resulting in dysregulated dendritic cell activation and low interaction with macrophages (Solana et al., 2014).

Gut microbiota
The gut microbiota refers to a significant number of bacteria and other microorganisms that inhabit the human digestive tract (Calder et al., 2022). "Gut microbiome" appeared in recent citation bursts, suggesting a likely current study emphasis (Fig. 10B). Many age-related changes, such as immune system dysregulation and disease vulnerability, are linked to gut microbiota (Bana and Cabreiro, 2019;Clemente et al., 2012). Aging has physiological impacts on both the host and the microbiota, and interactions between the host and microbiota may influence aging as a whole (Bana and Cabreiro, 2019). The distribution and diversity of the gut microbiota changes with age. Commensal microbes like bacteriodes, bifidobacteria, and lactobacilli are decreased, whereas opportunist microbes like enterobacteria, Clostridia perfringens, and Clostridia difficile are elevated. (Rea et al., 2012). These changes in gut microbiota may influence inflammaging due to constant stimulation of the immune system, which causes immunosenescence (Lakshminarayanan et al., 2014;Santoro et al., 2020). Overall, this inflammatory milieu contributes to the evolution of many clinical disorders in elderly individuals and increases the susceptibility of the host to harmful germs (Santoro et al., 2021;Bischoff, 2016).
Diet, area, and nation of residence (Ghosh et al., 2020), physical activity (Huang et al., 2019), and medications (Sun et al., 2019) are important factors that can regulate gut microbiota throughout life. Notably, centenarians (99-104 years old) and semi-supercentenarians (105-109 years old) demonstrate distinctive microbiota modifications, such as higher diversity, suggesting a close relationship between greater lifespan and the microbiota (Biagi et al., 2016;Wilmanski et al., 2021). It is unclear whether age-related alterations cause immunosenescence; however, restoring age-related gut microbial richness and function through tailored nutrition or supplementation may prevent a decline in immune fitness.

Table 4
The top 10 journals and co-cited journals that published papers on immunosenescence.    -Sowa et al., 2022). Alzheimer's disease (AD) and Parkinson's disease (PD) are the most prevalent neurodegenerative disorders associated with aging, and inflammation aging was shown to influence the decline in cognitive functioning and onset of dementia (Barbe-Tuana et al., 2020). In AD, microglia-mediated inflammation contributes to degenerative processes. Microglia produce proinflammatory cytokines and their ability to phagocytose amyloid-β plaques is compromised . Alterations in the immune system may play a role in PD development. Williams-Gray et al. (2018) discovered that peripheral immunological profiles of PD patients were uncommon for older people due to a lack of CD8+ T cell replicative senescence-a characteristic of normal aging. Patients with MS exhibit early thymic involution and diminished immunological functioning (Dema et al., 2021). Involution of the thymus reduces the generation of naive T-cells and T-cell activity (Vaughn et al., 2019). Eschborn et al. (2021) proposed that aging in MS was related with a loss of equilibrium between costimulatory and immunoregulatory signals provided by CD8+ T cells, favoring a proinflammatory phenotype. In addition, research indicates that senescent alterations in multiple retinal and choroidal tissue cells, concurrent with systemic immune aging, are significant contributors to the initiation and progression of age-related macular degeneration (Lee et al., 2021). Thus, immunosenescence is an important factor in determining the progression of neurodegenerative diseases, but their exact relationship needs further study.

Immunosenescence and rheumatoid arthritis.
Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by damage to the joints and other tissues. RA has been viewed as a paradigm of premature senescence, and its prevalence increases considerably with age (Doran et al., 2002). Several characteristics of immunosenescence have been observed in RA patients, including thymic dysfunction, clonal expansion of peripheral T cells, telomeric attrition, and production of proinflammatory cytokines (Bauer, 2020). A considerable expansion of late-differentiated or aged T lymphocytes (CD4+CD28-and CD8+CD28-) in RA was discovered (Petersen et al., 2015;Pawlik et al., 2003), similar to what has been found in healthy aging (Effros et al., 2005). Intriguingly, patients with extra-articular symptoms of RA exhibit a high proportion of these aged T cells (Martens et al., 1997). For instance, CD4+CD28-T cells are more abundant in RA patients with extraarticular inflammation or atherosclerotic illnesses, and the expansion of aged CD8+CD28-T cells is inversely linked with memory skills in RA patients (Petersen et al., 2015;Chalan et al., 2015). Thus, immunosenescence may exacerbate articular and extra-articular symptoms of RA.
The consequences of T cell senescence on the incidence and progression of RA have received considerable attention, indicating that immunosenescence has a negative impact on autoimmune disorders.

Immunosenescence and cancers.
In general, the occurrence of malignant tumors increases with age (Lian et al., 2020). Age-related increases in carcinogenesis are influenced by several mechanisms, including the growth of senescent cells, inflammaging, and immunosenescence changes linked to poor immune surveillance (Lian et al., 2020;Barbe-Tuana et al., 2020). In the elderly, immune function is disordered, resulting in an increase in tumor-infiltrating Treg cells that promote tumor development and metastasis (Singer et al., 2018). Immunosenescence, particularly in CD8+ T cells, plays a crucial role in the development and treatment of breast cancer (Onyema et al., 2015). In a mouse model of breast cancer, decreased IFN signaling in CD8+ T cells of aged mice was observed (Sceneay et al., 2019). Inflammation may also increase the risk of cancer associated with aging. Through tumor initiation, propagation, and progression, the inflammatory environment appears to support cancer development. Although several chemotherapy treatments induce senescence in cancer, the SASP can have a negative impact on cancer treatment (Barbe-Tuana et al., 2020). Additionally, persistent chronic infections such as cytomegalovirus have been linked to rapid immunosenescence and may contribute to poor cancer immunity (Bauer and Fuente, 2016).

Immunosenescence and COVID-19.
During the recent crown pandemic, the connection between immunosenescence and infection has garnered significant attention. Mortality due to COVID-19 increases with age and is especially high among the oldest population (Marcon et al., 2020). It is now known that COVID-19 patients with pre-existing conditions like cardiovascular disease, type 2 diabetes mellitus, chronic respiratory disease, arterial hypertension, and various cancers have a much higher mortality rate than those without any pre-existing conditions (Damayanthi et al., 2021). Direct SARS-CoV-2 infection of T cells is hypothesized to cause lymphopenia in patients with severe COVID-19 . This observation is critical for understanding how SARS-CoV-2 infection affects the aging immune system. COVID-19 can  induce CD4+ T cells to develop into pathogenic Th1 cells and generate proinflammatory cytokines, thereby causing a cytokine storm. Indeed, older individuals are more likely to develop a cytokine storm and its deadly effects than younger patients due to aging-related traits like mitochondrial dysfunction and reactive oxygen species (Wissler Gerdes et al., 2022;Rydyznski Moderbacher et al., 2020;Farshbafnadi et al., 2021). Thus, compromised immune activity in the elderly makes it easier for the virus to propagate and destroy tissues, hence highlighting the need to combat immunosenescence and enhance immunological function (through vaccination) to protect the body from COVID-19.

Possible therapeutic strategies
Currently, there are a variety of treatments for senescence and senescent cells. One of the primary characteristics of aging is the involution of the thymus, which may cause a decline in T cells; therefore, restoring the structure and function of the aging thymus has the potential to reverse immunosenescence (Kim et al., 2015). Drugs that use IL-7 as a naive T cell growth factor, checkpoint inhibitors that improve T cell responses, and mitogen-activated protein kinase inhibitors may be effective treatments (Aiello et al., 2019;Palacios-Pedrero et al., 2021). Clearing senescent cells is crucial because they recruit the SASP, and immunotherapy is an effective method of treating senescent cells. For example, engineered T-cells that express a chimeric antigen receptor have been successfully used in cancer therapy (June et al., 2018). Supplementation with exogenous calreticulin restored macrophage clearance of apoptotic cells lacking the cyclin-dependent kinase inhibitor 2 B (Kojima et al., 2014). In addition, it has been demonstrated that exercise decreases the frequency of Th17 cells and inflammatory markers while increasing IL-7, thymic function, and autophagy activity (Zhou et al., 2021). Finally, diets may affect aging by altering gut microbiota (Rodrigues et al., 2021). Thus, lifestyle also plays a major role in health and longevity and may result in a healthier aging process.

Strengths and limitations
Compared to a typical overview, this bibliometric study provides a more accurate depiction of ongoing research goals and trends, as well as a relatively exhaustive and objective data analysis. This study has some limitations. First, the cutoff date for the articles included in the analysis was December 31, 2021; thus, the current 2022 literature was not accounted for in our study. Second, this research contained both articles and reviews, and the disparate quality of the gathered literature could In this network map, all of the keywords could be classified into three clusters: cluster 1 (red nodes), cluster 2 (green nodes), and cluster 3 (blue nodes) (blue nodes). The node and word sizes reflect the frequency of co-occurrence, the connection shows co-occurrence, and the node color denotes the cluster. (B) Density visualization of keywords. Word size, circle size, and red opacity are correlated with higher cooccurrence frequencies. (C) Overlay map of co-occurrence keywords. The nodes indicated with purple or blue color represent earlier-appearing keywords, while the keywords marked with yellow color show current research concerns. (D) Three-field plot of the Keywords Plus analysis on immunosenescence. Middle field: authors; left field: keywords; right field: journals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) diminish the reliability of the map analysis. Third, we retrieved only English-language articles from the WoSCC database, eliminating those that were neither in WoSCC nor in English. However, literature-based visual analysis makes it easy for academics to understand study topics, hotspots for research, and development trends in immunosenescence.

Conclusions
Since 1970, there has been stable growth in immunosenescence research, with active global cooperation. Through comprehensive bibliometric analysis and the production of a visual map, this study summarized and analyzed the academic publication, research subjects, research hotspots, and development trends of immunosenescence research to provide a credible reference for the field. The United States was the global leader in this study. The University of California was the most frequently published among the research institutions. Pawelec G contributed the most to the publications, and Franceschi C was the most co-cited author. In addition, most articles on immunosenescence were published in important international journals with high IFs. Finally, "inflammaging", "gut microbiota", "cellular senescence", and "COVID-19" were hot topics in this field. Thus, future research should focus on the role of immunosenescence in age-related diseases, individualized treatment, and ways to reduce the negative effects of immunosenescence to promote health and longevity.

Declaration of competing interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability statement
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.