Circadian rhythm in cardiovascular diseases: a bibliometric analysis of the past, present, and future

All publications were from the Web of Science Core Collection (WoSCC). The data from WoSCC have the most complete data structure, including publication type (PT), author (AU), journal (SO), keyword (DE), abstract (AB), institution (CI) and reference (CR). The study used a search formula of “TS = (“Circadian Rhythm” OR “Diurnal Rhythm” OR “Twenty Four Hour Rhythm” OR “Circadian Clocks” OR “Clock System”) AND TS = (cardiovascular OR heart)”, which is described in Fig. 1 along with the inclusion criteria for the study. The following were the criteria for the publication’s inclusion: (1) research on circadian rhythm in CVD; (2) the type of publications included articles and reviews and freely available data; (3) the language of the publication was English. The following were the criteria for the publication’s exclusion: (1) the publications did not address the topic of the study; (2) the publications were news, conference abstracts, or briefs. To ensure accurate data updates, all of the aforementioned operations were completed within a 24-h period, on May 15th, 2022.

The document types were only articles and reviews. Duplicate studies were removed. All the information, including the number of papers and citations, publication year, titles, authors, affiliations, keywords, countries, journals, and references, was collected for bibliometric analysis.

In this study, VOSviewer (version 1.6.18), HistCite (version 12.03.17), and CiteSpace (version 5.8.R3) were used to perform the bibliometric analysis.

VOSviewer was used to visualize complex co-citation networks [34], such as the collaboration and time trends among countries, institutions, and individuals. The size of the nodes represents the number of publications; the thickness of the line represents the strength of the link, and the colors of the nodes represent different clusters or times.

HistCite was utilized to determine the total number of publication records, total global citation score (TGCS), and total local citation score (TLCS) for each publishing year, active nations, top institutions, core journals, and authors. The TLCS refers to the number of times an article is cited in the current database, which means attracting attention from the same field. Therefore, TLCS is used to rank the authors and journals. More crucially, it was used to find the sample citation routes of relevant references [35].

CiteSpace was used to assess the visual study of the knowledge domain and developing trends [36], such as cluster analysis, dual-map citation overlay, timeline or time zone views, references, and keywords citation bursts [37]. Cluster analysis may be used to categorize references and keywords as well as highlight interesting study topics on circadian rhythmicity in CVD. In cluster analysis, the modularity Q and mean silhouette are two key assessment metrics. Q greater than 0.3 implies that the clustering structure is substantial. The presence of a mean silhouette greater than 0.5 shows that the clustering results are credible. Keywords and references bursts are frequently employed to discover new research trends in the subject [38]. It has two interpretations: (1) in keyword analysis, it represents the frequency of words or phrases used in the citing document. (2) In co-cited references analysis, it represents the frequency of citations received by the cited document.

A total of 2102 publications related to the circadian rhythm in CVD were retrieved from WoSCC. Linear regression analysis showed that the annual number of publications on circadian rhythm in CVD has undergone an overall increasing trend since 2002 (Fig. 2). So far, the annual growth trend is in line with the fitting curve y = 0.0095x⁴ − 0.342x³ + 4.1389x² − 15.065x + 76.526 (R² = 0.9552). According to this curve, it is predicted that the number of publications in 2022 will reach about 250. These publications have been cited 62,070 times, with an average of 29.53 times per publication. The annual publication count exhibits fluctuations, however, on the whole, it is increasing and thus indicates promising prospects for research in this area.

From 2002 to 2022, 80 countries and regions published research papers on circadian rhythm in CVD. The shade of green reflected the amount of countries’ publications—the darker the green, the greater the number of publication (Fig. 3A). The top 10 countries with the highest number of publications have generated about 74.59% (n = 1568) of the publications in the world (Fig. 3B). The United States was the most cited country for published articles (cited 28,803 times), followed by Japan (cited 7258 times), and Italy (cited 5599 times). The top three countries in centrality were the United States (centrality = 0.72), China (centrality = 0.20), and the UK (centrality = 0.19), demonstrating these countries contribution in this field (Table 1). The international collaboration network visualization showed that the collaboration between countries was relatively close. The United States has been in collaboration with almost all the other countries (Fig. 3C). From the perspective of the publication number, the USA has consistently been the top publisher (biggest node), Japan has been stable (deepest color), and China is a late starter in this field compared to other countries, but it is developing rapidly (Fig. 3D).

The top 10 institutions with the highest output on circadian rhythm in CVD research are shown in Table 2. Harvard University (n = 39) was the leading institution in terms of output, followed by the University of Tokyo (n = 35), the University of Alabama at Birmingham (n = 33), and the University of Ferrara (n = 31). Institutional collaboration was relatively high level. The Harvard University-led collaboration groups demonstrated the closest collaboration with other schools (Fig. 4A).

The top three most productive authors were Hermida RC (published 34 articles), Young ME (published 25 articles), and Ayala DE (published 24 articles). In Table 3, Hermida RC of Universidade de Vigo got the highest TLCS (score = 398), followed by Ayala DE of Universidade de Vigo (score = 287) and Portaluppi F of the University of Ferrara (score = 271).

Through TLCS data, we discovered that the study of Universidade de Vigo and its researchers piqued the interest of a large number of scholars. But Fig. 4A shows that the link lines between Universidade de Vigo and other universities were insufficient and sparse, implying that Universidade de Vigo lacked collaboration with others. This result seems inharmonious, implying that there may be some concealed bias in Vigo University’s work. Their work, for example, may lack independent source data validation from non-host universities [39]. The visualization analysis of the author is shown in Fig. 4B. It is apparent that collaboration between authors from other institutions is lacking, and they prefer to collaborate inside their affiliations.

856 journals published the articles on circadian rhythm in CVD. The top 10 journals with the highest TLCS are shown in Table 4. About 14.41% (n = 303) of the articles were published in these journals. The Chronobiology International (TLCS = 467) was the most prolific journal, followed by the Circulation (TLCS = 327) and the Circulation Research (TLCS = 226). Four main citation pathways were depicted in the dual-map overlay, the right for cited journals and the left for citing journals (the journal in which a source article is published is called a “citing journal”. A reference cited by a source article is called a “cited article”) (Fig. 5). The dual-map provides an understanding of the past development direction of the field of circadian rhythm in CVD [40]. The citing publications were mostly published in journals in the field of molecular biology, immunology, medicine, and medical clinical, whereas most of the cited publications were published in journals in the field of molecular biology, genetics, health nursing, and medicine.

The article’s basic content comprised keywords. Keyword analysis can be used to investigate research hotspots and frontiers in a field. During data collection, we collected 6548 keywords, and a cluster analysis produced 6 clustering outcomes (Table 5). The co-occurrence analysis of keywords revealed the primary topic in the field of circadian rhythm in CVD by presenting the keywords with a high frequency. The “circadian rhythm” was the most frequent keyword and held the central position in the network map (Fig. 6A). Figure 6B displays a screenshot of the timeline view, showcasing clusters arranged horizontally along timelines. The label for each cluster is positioned at the end of its corresponding timeline. The “cardiac troponin” (mean year = 2013) and the “stress-induced change” was the earliest (mean year = 2006) are the latest and earliest subjects, respectively. Moreover, burst keywords have been detected, which refers to keywords with a strong increase in frequency. The “strength” represents the burst extent. The “Begin” and “End” described the duration of the surge. A total of 82 keywords were extracted by keyword burst analysis, the top 25 of which are shown in Fig. 6C. The keyword “oxidative stress” had the highest burst strength in recent years.

When two publications were cited by a third document at the same time, the two documents are said to have a co-citation relationship; the higher the co-citation frequency, the closer their academic link and “distance”. We classify the publications in the discipline field based on this "distance" using the statistical approach of cluster analysis and visually identify and assess the segmentation in the discipline field using a graphical representation. The top 10 highest TLCS references included 2 reviews and 8 research articles (Table 6). Storch et al., authors of the article with the highest TLCS, found through a comparative analysis of mouse liver and heart in vivo that peripheral circadian gene regulation is extensive. They discovered marked differences in the distribution of circadian phases between the two tissues and only a few genes with circadian regulation in both [41]. Citespace was utilized to construct a visual network of co-cited references and a cluster analysis of documents was conducted. A total of 15 clusters were found, the modularity Q was 0.7951, and the mean silhouette value was 0.9294. There were nine clusters with the highest size values found (Table 7; Fig. 7A), which include “inflammation”, “endothelium”, and “myocardial infarction”, among others. We also designed a timeline view for clusters (Fig. 7B). We found that “hormones” is an early field of circadian rhythm in CVD (mean year = 1998). However, “inflammation” and “myocardial infarction” are the current hot topics (mean year = 2016/2014). Finally, a reference burst was performed from the co-citation investigation by Citespace. Figure 7C displays the top 25 references with the strongest citation bursts, which shows the most representative references in terms of burst strength, burst duration, and burst time. The team of Morris (2016) exhibited the highest burst strength. In their article, they demonstrated that circadian misalignment alone increases blood pressure and inflammatory markers [42].