Worldwide research trends and key focus areas on PCSK9 and heart disease

Bibliometrics is a method of analyzing publications from quantitative and qualitative perspectives to determine their output and status. It was formed as a separate discipline in 1969 (42) and was widely used in literature analysis (43). Devey summarized that “bibliometrics is to scientific papers what epidemiology is to patients” (44). It can help researchers quickly understand the trajectory of a topic by tracking its spread through the literature. It can also characterize actively and heavily downloaded journal articles to assess their impacts (45).

Through the analysis process, we can obtain detailed information concerning authors, keywords, journals, countries, institutions, and references in a specific field, enabling the unveiling of the field's development through bibliometric analysis (46). The visualization can provide an intuitive presentation and a good complement to the results of bibliometric analysis. Furthermore, visualization enables the exploration of connections among information, such as similarities in research topics among different authors, the research focus of various institutions, and the emergence of new theories from existing institutions. The common tools include VOSviwer, CiteSpace, etc.

CiteSpace is a web-based Java application that is unique and influential in the field of information visualization and analysis. It provides a visual representation of trends in a discipline or field of knowledge, showing how they have evolved over time. This is done by using similarity algorithms to analyze the literature and study the development of various research frontiers. Therefore, the temporal dimension can be employed to clearly outline the evolution of knowledge and the historical scope of literature within a specific cluster. This elucidates the developmental process and trajectory of the field (47).

VOSviewer was released in 2010 by Nees Jan van Eck and Ludo Waltman (48). It is a software tool for creating and exploring maps based on web data. VOSviewer’s advantages are easy to operate and direct aesthetics on visualization (49).

The full documents and citation files from the Web of Science searches are stored in plain text (txt) for analysis. They are then imported into VOSviewer and CiteSpace as a dataset. These tools are utilized for the analysis and creation of visualization maps. In addition, the bibliometrics online analysis website and Prism8 are employed for graphical mapping in this study.

This study retrieved relevant literature from the Web of Science Core Collection database (WOSCC). WOSCC is widely regarded as one of the world's most authoritative, systematic, and comprehensive databases. This is attributed to its extensive research in biomedical sciences and its superior capability to track older citations compared to Scopus. WOSCC is also recognized for more precise categorization of journals than Scopus (50). The literature search was completed in one day on July 1, 2023 to minimize the impact of frequent updates to the bibliographic database. Science Citation Index Expanded (SCI-expanded) and Social Science Citation Index (SSCI) were used as data sources. We established our search strategy as “TS = [(“cardiovascular disease*” OR “cardiac vascular disease” OR “coronary heart disease” OR “cardiovascular events” OR “coronary artery disease” OR “cardiovascular risk”) AND (“PCSK9”)]”. Literature types included regular and review articles. Time span was between 2006-01-01 and 2023-07-01. The language of publications was restricted to English. The articles were exported and stored in txt format (including full records and cited references) for further analysis (Table 1).

The information on publications was arranged in descending order of their publishing time. To assess the suitability of the studies, they were screened for title and abstract separately. The researchers whose titles and abstracts met the inclusion criteria were included in the full text for searching and evaluation. Figure 1 displays the screening process.

After the screening, we obtained a cluster network containing 2,474 articles. The general information, trends, and focus of new releases in the field of cardiovascular diseases and PCSK9 over the last 18 years were analyzed using bibliometric analysis software (VOSviewer 1.6.19 and CiteSpace 5.7.R5), online bibliometric website (bibliometric.com↗) and Prism8. Performance analysis and visualization of authors, countries, institutions, journal distributions, most impactful literature, and keywords were conducted.

After de-duplication, a total of 2,474 relevant research papers on PCSK9 and CVD were included, with 1,603 academic papers (accounting for 64.8%) and 871 review (accounting for 35.2%). These researches came from 10,323 authors in 86 countries and 3,243 institutions. These articles were cited a total of 98,330 times. The average number of citations reached 37.73 and were published in 590 journals. 51,003 references from 6,750 journals were cited. Since Cohen, J.C. published the first article in 2006. The number of publications in this field increased from 5 in 2006 to 79 in 2014. The literatures increase year by year. The growth rate of the publications reached 93.7% from 2014–2015 and reached a peak in 2019, with a total of 300 publications this year. More than 250 of researches were published from 2018–2022 (Figure 2). The number of citations increased year by year from 2006–2022 and reached its highest level in 2022. This indicates that the field is maturing and entering a plateau. The detailed numbers of annual papers are presented in Figure 2.

We have mapped network maps using VOSviewer and SCImago Graphica software to understand the top contributing countries and regions and the extensive collaborations among them. The visualized network maps show the numbers of publications and collaborations among countries. The node size indicates the number of published literatures, while the line thickness represents the strength of the links among countries or institutions. The analysis of international cooperation based on 57 countries was displayed in Figures 3A,B. We discover the geographical distribution of publications covering 86 countries/regions on six continents. The circular map depicts the number of publications, with larger sizes indicating a higher number of publications. Links represent the connections among countries or institutions, with thicker links implying more collaborations. Among these countries, 12 have published only one article, while 57 countries have published at least five articles. According to Table 2, the ten most productive countries are the USA (936 papers), England (313), Italy (308), China (302), and Canada (252). The top five countries that work most closely with China are the United Kingdom (link strength of 160), Canada (127), France (121), Netherlands (112), and Germany (88). The United States stands out as the most collaborative country with China, with a link strength of 43 (Figure 3A).

An analysis of the number of citations for countries shows that the United States has the most citations, followed by the United Kingdom, Canada, France, and the Netherlands. In terms of the average number of citations per article, the Netherlands has the highest with 101.17, followed by Australia with 96.45, England (87.41), Canada (84.22), and France (83.92).

Through our analysis, we found a total of 3,243 institutions involved in the research on cardiovascular diseases and PCSK9. The top five most prolific institutions were Amgen Inc (114 articles), Harvard Med Sch (96 articles), Sanofi (94 articles), Univ Milan (86 articles), and Imperial Coll London (78 articles). The top ten institutions in terms of contribution and centrality are shown in Table 3. The centrality is usually used to indicate the connectivity of each node. The node with a high mid-centrality indicates that it acts as a key point connecting two or more groups displaying transformation patterns (42). Through CiteSpace software analysis, we found that the top five highest centrality values were Univ Amsterdam (0.17), Capital Med Univ (0.14), Imperial Coll London (0.14), Brigham & Womens Hosp (0.13), and Baylor Coll Med (0.13). To explore when institutional research began and continued, we constructed a timeline graph about institutional research through the visualization feature of VOSviewer. In Figure 4A, the circular nodes size shows the number of publications. The colors of the nodes show the average publications in the first published year in the field. Montreal and Heart Res Inst. started to research earlier, while Mediterranea Cardioctr, Univ Messina, and Chinese academic med sci&peking have conducted more recent research in this field. Figure 4B presents the inter-institutional collaborations. The connecting lines of the nodes represent their collaborations. The intensity of the lines represents the closeness of the collaboration. The institutions that collaborate most closely are Univ Milan and Irccs Multimed, Univ Western Australia and Royal Perth Hosp, both with relative connecting line intensities of 36.

We analyzed the results using VOSviewer and CiteSpace to understand the most influential and active authors in the field. The analysis revealed that a total of 10,323 authors contributed to all the 2,474 papers. The number of authors for a single paper, also known as transient indexing, was 8,098, accounting for 78.45% of all authors. Notably, 143 authors have published 10 or more papers in this field. The most prolific author in the field is Gerald F. Watts, with 49 publications and an h-index of 97. Secondly is Marc S. Sabatine with 48 papers (h-index 117) and Robert P. Giugliano with 44 papers (h-index 88) (Figure 5A).

By analyzing Gerald F. Watts' articles, we found that his interests include the diagnosis and treatment of familial hypercholesterolemia, evidence for LDL-induced cardiovascular diseases, and treatment guidelines. He published a total of 49 papers from 2006 to February 2022, receiving a total of 2,006 citations, with an average of 40.94 citations per paper. Gerald F. Watts is affiliated with the Royal Perth Hospital.

Marc S. Sabatine, who ranked second in terms of publication productivity, published 48 papers with a total of 9,376 citations, averaging 195.3 citations per paper. He served at Harvard Medical School and focused on the evolocumab and clinical outcomes in patients with cardiovascular diseases, dapagliflozin, and cardiovascular risk in type 2 diabetes mellitus. There is a close collaboration between Marc S. Sabatine and Gerald F. Watts in the field, particularly in the study of evolocumab's efficacy and safety in reducing blood lipids and cardiovascular events.

When analyzing citations, we found a total of 9,325 authors who had one or more citations, accounting for 90.33% of the total 10,323 authors. Among them, 5,965 authors had 10 or more citations (57.78%), and 1,195 authors had 100 or more citations (11.58%). The most influential authors were Wasserman, Scott M. (h-index 55), Sabatine, Marc S. (h-index 117), and Giugliano, Robert P. (h-index 88). Wasserman published a total of 42 papers, receiving 10,504 citations with an average of 250.1 citations per paper. Wasserman's focused on evolocumab (AMG145) for lipid-lowering, a PCSK9 inhibitor, as well as its efficacy and safety for cardiovascular events and PCSK9 in familial hypercholesterolemia. This author also had a strong collaboration with Marc S. Sabatine in this field. Among the co-cited authors, the most co-cited author is Marc S. Sabatine (h-index 117) with 1,677 citations, followed by F.J. Raal (h-index 58) with 1,236 citations (Figure 6). Additionally, we constructed a network of co-citations (Figure 5B).

In terms of author collaboration, we found that out of the 143 authors who published 10 or more papers. Most of the authors formed only a few collaborative groups. Each group revolved around collaboration with one or two highly prolific authors, as shown in the author's collaboration visualization diagram (Figure 5C). Notably, there were relatively few connections between different groups, suggesting a lack of collaborations between research teams/laboratories in the researches related to cardiovascular diseases and PCSK9.

Two of the most widely used methods to assess the quality of the papers in medical journals are the number of citations and the impact factor (IF). The impact factor (IF) is widely recognized as one of the main indicators that primarily evaluates the quality, importance, and impact of medical journals in their respective disciplines (51, 52). In this research, we found that articles on cardiovascular diseases and PCSK9 were published in 590 journals. Atherosclerosis has the highest publications at 111. We then analyzed the recent impact factors (IF) of the top 10 journals with the highest number of publications. Among these journals, Circulation had the highest IF (Table 4). Additionally, we research the publication volume of these top 10 journals by using Journal Citation Reports (JCR) and found that the majority (60%) were categorized as Q1. In terms of the journal category, the heart and cardiovascular system accounted for 60% of the journal categorization. Finally, when it came to the place of publication among the top 10 journals, among those journals, 7 were in the United States, 2 were in the United Kingdom, and 1 was in the Netherlands.

The journal dual graph overlay function of CiteSpace was used to construct the disciplinary distribution of academic journals, presented in Figure 7. The area covered by the citing journals is shown on the left-hand side, while the area covered by the cited journals is shown on another side. The dual-map overlay of journals shows 2 major citation paths. The orange path is an independent development pattern, the citing journals mainly belong to MOLECULAR, BIOLOGY, and IMMUNOLOGY, and cited journals mainly belong to TEHNOLOGIJE, METALURGIJA, MIDEM-JOURNAL and MOLECULAR, BIOLOGY, GENETICS. This path indicated that the field has evolved independently from the cited research background topics such as TEHNOLOGIJE, METALURGIJA to the research frontier topics such as MOLECULAR, IMMUNOLOGY. Whereas, the green pathway is a confluent pattern of development, indicating that the two cross-cutting areas evolved into a common research theme, with DERMATOLOGY, DENTISTRY, SURGERY and HEALTH, NURSING, MEDICINE, and the cited themes such as TEHNOLOGIJE, METALURGIJA, MIDEM-JOURNAL, etc. The themes together provide the research basis for the development of the frontiers of NEUROLOGY, SPORTS, OPHTHALMOLOGY, MEDICINE, MEDICAL and CLINICAL. Thus, understanding the development process of the field can provide a good knowledge reference direction for the current research frontiers and grasp the development pulse of the researches.

Citation frequency analysis is a valuable method for evaluating highly cited papers. Citation frequency can reflect the impact of an article in a particular research field (53). In order to explore the most influential research in the field, we analyzed the citation frequency of articles by VOSviewer. A co-occurrence visualization network of article citation counts was derived. The ten most cited articles were published between 2006 and 2019, and all of those were cited more than 1,000 times. Half of these articles were published in the New England Journal, two in Nature Genetics, and three in cardiology journals in the United States and Europe. The first authors of the ten highly cited articles were all from the United States. The cited journals' co-occurrence and top 10 cited documents are charted in Figures 7, 8A.

Keywords unite the core and essence of an article. Keywords co-occurrence can identify research hotspots in the field. We analyzed the research hotspots in the literatures by using VOSviewer, analyzing the keyword co-occurrence network and density view in this field. A total of 5,600 keywords, the most frequently occurring keywords were PCSK9 (903 times), cardiovascular disease (804 times), statin therapy (790 times), familial hypercholesterolemia (787 times), PCSK9 inhibitor (539 times), and cholesterol (509 times). The keyword density view is presented in Figure 8C. The more reddish color indicates the higher frequency of keywords.

Clustering keywords can reveal the structure of interconnected research fields. The keyword co-occurrence network mapping analysis was performed on the keyword documents with the top one hundred frequency of occurrence (Figure 8B), in which the nodes that are close to each other and have the same color can be regarded as the same cluster, and three clusters are demonstrated in Figure 8B. Cluster one is in red and contains 42 keywords, including: PCSK9, cardiovascular disease, familial hypercholesterolemia, atherosclerosis, gene expression, mutation, cholesterol, LDLR, etc. Cluster II is in green color and contains 35 keywords, including: PCSK9 inhibitor, evolocumab, alirocumab, statin therapy, cardiovascular risk, safety, efficiency, and so on. Cluster III is in blue and contains 18 keywords, including: statin/kexin type9, double bind, statin tolerance patience, monoclonal antibody, evolocumab amg145, heterozygous familiar hypercholesterolemia, etc.

In order to understand the time point when the keywords appeared and received widespread attention, further analysis was conducted through CiteSpace software and the online bibliometric website. The timeline of the keywords from 2006–2023 is presented in Figure 8D. The more the line color leans towards red means that the keyword has received widespread attention more than recently. While the more it leans towards purple means that the keyword has appeared earlier, and the dotted line means that the keyword has not received much attention in this period. A river diagram about the publication of PCSK9-related topics was presented in Figure 8E, which is based on the time series, counting the number of papers year by year according to different technical subject terms, and forming a trend change diagram within a certain time range. Demonstrating the distribution of PCSK9 in the time dimension can visualize the proportion and change of each related topic in the whole, and also helps to discover potential patterns and associations.

Detecting outbreak words is crucial for identifying evolving research hotspots in an academic field. High citation bursts for articles or keywords suggest that they were widely discussed or utilized during a specific period. To reveal the fastest-growing topics in recent years, we performed keyword outbreak analysis via CiteSpace, Figure 8F presents the 25 keywords with the highest outbreak intensity. The blue line shows the time interval observed from 2006–2023 while the red line shows the duration of the outbreak. The highest outbreak intensity of keyword is monoclonal antibody. Its outbreak intensity is 40.17, which is certified this keyword became a hotspot in 2016 in this field. According to the placebo-controlled trials, autosomal dominant hypercholesterolemia, mutation, LDL receptor, and other keywords are the top 5 highest outbreak intensity and last for many years. Researchers show continuous interest in it. The latest keyword outbreaks include proprotein convertase subtilisin, inflammation, bempedoic acid, and inclisiran, which could be potential frontiers in the field of cardiovascular disease and PCSK9 research.

Starting with the first publication in this field in 2006, there was an exploratory start-up process for the following decade. There was a steady increase in document publication number and the extension of document types after 2015. This growth continued after 2018. About 300 articles were published per year. The result shows that the field of research has entered a period of stabilization. The reason for the large increase in articles in 2015 was probably due to the initial introduction of PCSK9 antibodies (23). Monoclonal antibodies targeting PCSK9 were rapidly developed and studied in large clinical programs over the following years. By 2018, three cardiovascular outcome trials demonstrated that PCSK9 inhibitors significantly reduced the risk of major vascular events (13). Even in the context of already being treated with maximal doses of statins, PCSK9 inhibitors lowered plasma LDL-C levels by around 60%, PCSK9 inhibitors, unlike statins, do not result in serious adverse effects, including excessive myalgia, elevated plasma hepatic aminotransferase levels, diabetes mellitus, or neurocognitive adverse events (13). Furthermore, a study (54) showed that evolocumab could be safely and well-tolerated over a median follow-up period of more than seven years. Early intensive use of evolocumab for LDL-C lowering results in a sustained cumulative effect of cardiovascular benefit. To maximize the clinical benefit, it is suggested that early initiation and sustained lowering of LDL-C should be prioritized (54). These trials proved that the PCSK9 inhibitors can reduce LDL-C levels to unprecedentedly low levels (55). It also suggested that more aggressive lipid-lowering strategies should be employed against LDL-C targets.

Inter-country collaboration has played an important role in the field. The United States, as the country with the highest number of articles, occupies a central position in global CVD and PCSK9 research and is at the forefront of this research field in terms of its influence and activity. There is extensive communication and cooperation with several other active countries. USA, Austrila, United Kindom and Canada are the leading countries in terms of publications and collaborative networks, suggesting that these countries may play an important bridging role in research cooperation. In terms of the quality of articles, Netherlands ranked first in terms of citations per article with a high average of 101.17, which indicates that the high quality of articles written by authors from this country is generally valued. On the contrary, China has the fourth highest number of publications but the lowest average of 14.58 citations per article. This highlights the need to prioritize the quality of articles rather than the quantity.

Co-authorship analysis is a useful method for identifying existing partnerships and facilitating the development of potential partnerships (56). Collaboration between researchers helps to explore existing collaborations and find potential collaborators. In the author co-authorship mapping, two highly cited authors, Sabatine, Marc S. and Wasserman, Scott M., have both focused on the field of evolocumab's efficacy and safety in reducing blood lipids and cardiovascular events. They have produced a number of classic works with top ten impacts in the field. The most cited article (Sabatine2017B) was contributed to them with a citation frequency of 3,114 times. This article was a randomized, double-blind, placebo-controlled trial involving 27,564 patients, the median duration of follow-up was 2.2 years. These findings show that patients with athero-sclerotic cardiovascular disease benefit from lowering of LDL cholesterol levels below current targets (55).

In this study, we analyzed the recent impact factor (IF) and journal citation report (JCR) of the top 10 journals with the highest number of publications. Most of the articles were still published in high-impact journals, with 60% classified as Q1. In terms of journal categories, the majority (60%) focused on the heart and cardiovascular systems. Regarding the publishers of the journals, the United States had seven of the top 10 journals, while the United Kingdom had 2 journals and the Netherlands had 1 journal. Notably, there were no high-impact journals in Asia on the list, emphasizing the need to develop influential international journals in the field in Asia.

By analyzing the institutions, we find that the most prolific institution is Amgen Inc., and the institution with the widest reach is Univ Amsterdam. More than half of the top ten activity institutions are universities, indicating that universities are the mainstay of the research field. For the research timeline, Clin Res inst., Montreal and Heart Res inst. are the pioneers in conducting research in this field. Meanwhile, Mediterranea cardiocyte, University Messina, and Chinese Acad Med sci&peking are currently engaged in exploring the latest advancements in this area. Inter-institutional partnerships play a vital role in academia, Univ Milan and Irccs Multimed Univ Western Australia, and Royal Perth Hosp are the institutions that collaborate most closely.

This study checked the top 10 most cited articles and identified their Web of Science categories and research directions, which were mainly in the cardiovascular system and cardiology, lipids, and the use of statins. More than half of these articles revealed that different antibody types of PCSK9 such as Evolocumab and Alirocumab can inhibit PCSK9 and significantly reduce LDL cholesterol levels and reduce the risk of cardiovascular events in the context of statin therapy. These studies also presented that patients with atherosclerotic cardiovascular diseases benefit from reducing LDL cholesterol levels below current target levels, which significantly reduces their risk of developing cardiovascular diseases, and analyzed dyslipidemia and cardiovascular diseases risk from a genetic perspective.

The keywords reflect the core theme and main content of articles, which can provide a reasonable description of the research hotspots. Through the keyword mapping analysis, we find that the prevention and treatment of cardiovascular diseases as well as the diagnosis and treatment of familial hypercholesterolemia are the main research hotspots. Furthermore, recent studies (57) have shown that pretreatment with PCSK9 inhibitor (PCSK9i), at the time of preischemic myocardial ischemia, can play a critical role in prevention of this disease. This is reflected in the reduction of myocardial infarction area and control of arrhythmia during myocardial I/R injury. The mechanism of cardio-protection may be related to the improvement of mitochondrial function and the phosphorylation of conjugated protein 43(Cx43), ultimately leading to improved left ventricle function. Inhibition of PCSK9 reduces the amount of reactive oxygen species (ROS), which result in attenuating mitochondrial damage, reducing mitochondrial swelling, and decreasing depolarization of the membrane potential. These effects provide further protection for the mitochondria. Additionally, the antiarrhythmic effects of PCSK9 inhibitors may be attributed to their interaction with Cx43 (57). However, further investigation is needed to understand the specific pathways involved in these mechanisms.

Researches on familial hypercholesterolemia are also one of the hotspots. PCSK9 is the third locus associated with familial hypercholesterolemia. Studies demonstrated that mutations in the PCSK9 caused autosomal dominant hypercholesterolaemia (58). PCSK9 binds to lipid-associated receptors to regulate metabolic mutations, leading to a higher plasma LDL-C clearance due to a reduction in hepatic LDL-R degradation. In addition, researches show that VLDL-R and ApoER2 are not only the most closely related family members of LDLR but also targets of PCSK9, whose main functions are involved in neuronal development and lipid metabolism (23). This leads to new measures for the diagnosis and prevention of familial hypercholesterolemia from a genetic perspective. Rare gain-of-function mutations in human PCSK9 cause familial hypercholesterolemia (58). Gene editing in vivo is an emerging therapeutic approach that allows DNA modification in patients (e.g., liver). CRISPR base editor is an attractive gene editing modality. The PCSK9 is a candidate target for in vivo gene editing (59). Liver-specific knockdown of this gene using the small interfering RNA (siRNA) inclisiran has been reported to produce a therapeutic effect on patients' lipid levels that lasts for several months (60). In principle, a one-time edit of PCSK9 might be able to be permanent to bring down blood levels of LDL cholesterol, thereby significantly reducing the risk of cumulative exposure to LDL cholesterol (61). These results may help to discover new therapeutic strategies against familial hypercholesterolemia.

According to keyword timeline and burst analysis, it is evident that proprotein convertase subtilisin, inflammation, bempedoic acid, and inclisiran may emerge as new potential frontiers in cardiovascular diseases and PCSK9 research. Although the primary action of PCSK9 inhibitors is mediated through the upregulation of LDLR and the subsequent dramatic reduction in circulating LDL-C levels, there is growing evidence that the PCSK9 may have pleiotropic effects. Inflammation has attracted the attention of scholars in recent years. One such role appears to be the modulation of inflammatory mechanisms in atherosclerotic cardiovascular diseases (ASCVD). PCSK9 may play an important role in SRscavenger receptors (SRs) expression, ox-LDL uptake, foam cell formation, and neointimal proliferation. Evidence indicates that PCSK9 is expressed in the infarct border zone. Therefore, targeted therapy against PCSK9, possibly in conjunction with inhibitors of other inflammatory mediators, may provide an effective therapeutic option to reduce the burden of ASCVD (62). This may be able to contribute to the development of new therapeutic strategies for PCSK9 application in the cardiovascular field as well as in other areas.

In summary, our study explores the journey of the global CVD and PCSK9 related fields from the emergence of research to the present. By analyzing these articles, we describe the major advancements in PCSK9 research regarding topics, institutions, countries, authors, and study types in cardiovascular disease research, offering researchers a quick understanding of the major research directions in this field. Our study also highlights the potential hotspots and frontiers of research.

In this study, the language of all articles included was limited to English. We may miss some high-impact articles related to CVD and PCSK9 but published in other languages. Second, we only searched the web of scientific databases. The type of publication only included regular and review articles. There were also limitations in the scope of literature source access. Thirdly, there may be contradictions in various aspects, such as an institution using different names at different times. Bias may result from several factors, including disproportionate or inappropriate citations due to institutional bias, linguistic bias, self-citation, bias of the powerful, internal bias, and intentional omission of citations for reasons such as competition (63, 64).