The retrieval was conducted on September 19, 2024. A keyword search with synonym expansion was performed using the Science Citation Index Expanded of the Web of Science Core Collection (WOSCC), covering the period from January 1, 2004, to September 19, 2024. To enhance accuracy in the bibliometric results, the search was conducted using a combination of title (TI), abstract (AB), and author keywords (AK). The search strategy was as follows: TI = (“patent foramen ovale closure” OR “PFO closure”) OR AB = (“patent foramen ovale closure” OR “PFO closure”) OR AK = (“patent foramen ovale closure” OR “PFO closure”). A total of 1,456 documents were retrieved. After excluding non-English publications and document types such as case reports, meeting abstracts, editorial materials, and other non-research articles, a final set of 927 English-language papers was obtained. This included 763 original research articles and 164 reviews. These articles were exported in plain text format and saved as “download_***.txt.”

Web of Science citation reports were used to calculate both annual and cumulative publication volumes. Bar charts illustrating trends in PFO closure research over the past 21 years were generated using Microsoft Excel (Office 2021).

A chart showing annual publication counts for the top 10 countries by research output was created using the online bibliometric analysis platform.¹

International collaboration among countries and regions was visualized using Bibliometrix R (4.4.1). Bradford’s Law was applied to identify core journals in the PFO closure field. Additionally, a thematic map was developed through keyword analysis.

Network visualization analyses of countries/regions, authors, and institutions were performed using VOSviewer (1.6.20). Both network and overlay visualizations were conducted for keywords. In these maps, the shorter the distance between nodes, the greater the similarity between the topics they represent (7).

CiteSpace (6.1. R1 Advanced) (8) was used to identify emerging research hotspots and monitor dynamic trends through burst detection of keywords and references. A timeline map of keyword clusters derived from co-cited literature was also generated using CiteSpace to enable in-depth temporal analysis.

A total of 927 PFO closure-related research papers published between January 1, 2004, and September 19, 2024, were retrieved from the WOSCC database, comprising 763 original articles and 164 reviews. Figure 1 presents the flowchart for literature screening and bibliometric analysis.

The number of publications increased steadily from 17 in 2004 to 74 in 2023 (Figure 2A). From 2004 to 2017, the growth was gradual, with annual outputs rarely exceeding 50. A notable rise occurred between 2018 and 2024, despite a slight dip in 2019 (<50 publications). In the past 7 years alone, 423 papers were published, averaging approximately 60 per year, reflecting a growing research interest in this field.

The top 10 countries/regions by publication volume are shown in Figure 2B. Over the 21-year period, the leading contributors were the United States (n = 321, 34.62%), Italy (n = 156, 16.82%), and Germany (n = 107, 11.54%). Although Chinese researchers entered the field later, their output surged after 2017. Over the past 3 years, China has surpassed all other countries in annual publication volume, highlighting its growing focus on PFO closure research.

A total of 56 countries have contributed to PFO closure research. International collaboration was analyzed using Bibliometrix R (Figure 3A). The United States leads in overall research output and collaborates most frequently with Germany (30 collaborations) and Switzerland (21 collaborations). The collaboration network in Figure 3B illustrates publication volume (represented by circle size) and partnership strength (represented by line thickness). Total Link Strength (TLS), which reflects the intensity of co-authorship, identifies the top five countries as follows: the United States (TLS = 213), Germany (TLS = 148), England (TLS = 140), Switzerland (TLS = 129), and Canada (TLS = 110). In terms of publication count, the top contributing countries are the United States, Italy, Germany, Switzerland, and China. The network map demonstrates active global cooperation, particularly between China and countries such as the United States, Germany, the UK, Canada, and Switzerland, as well as strong ties between the United States and several European nations.

Table 1 summarizes the top 10 countries/regions by number of publications, total citations, and h-index. The United States leads with over one-third of global output (accounting for 85.33% of publications among the top 10), the highest number of citations (11,178), and the highest h-index (51). Switzerland ranks second in citations (3,780) and has an h-index of 32.

A total of 1,393 institutions have contributed to research in this field. Among them, the top 10 institutions based on TLS collectively produced 230 papers. Of these, seven are based in the United States, two in Switzerland, and one in Canada (Figure 4A). Institutions in the United States and Switzerland are particularly prominent in terms of publication volume, with the University of Bern, the University of California, Los Angeles, and the University of Pennsylvania leading the field. To further analyze collaboration patterns, 209 institutions were selected for visualization, applying a threshold of at least three publications. A collaboration network was constructed based on the number of publications per institution and their inter-institutional relationships.

A total of 4,209 authors have contributed to PFO closure research. The co-authorship network (Figure 5A) illustrates collaborative relationships, where node size represents publication output and line thickness reflects the strength of collaboration. The five leading authors by TLS are: Meier Bernhard, Mattle Heinrich P., Rigatelli Gianluca, Windecker Stephan, and Kasner Scott E. These researchers have significantly advanced the field through extensive international collaboration. Co-citation analysis of 7,156 authors (Figure 5B) identified six authors who were co-cited more than 300 times: Mas JL (598 citations), Meier B (410), Homma S (345), and Kent DM (334). The network shows strong links between key figures such as Mas JL, Kent DM, and Homma S., emphasizing their foundational contributions to the field. Timeline clustering of co-cited references (Figure 5C) revealed evolving research trends. Earlier studies focused on single-center experiences and right-to-left shunts, whereas more recent work highlights atrial fibrillation and clinical characteristics. Sudden detection refers to identifying a sharp rise in citation frequency of a specific topic, article, author, or journal over a defined time period (9). Citation burst analysis identified 25 high-impact publications (Figure 5D). The strongest burst (64.34) was for the 2017 New England Journal of Medicine (NEJM) study by Søndergaard et al., which demonstrated a reduced risk of recurrent stroke with PFO closure plus antiplatelet therapy compared to antiplatelet therapy alone, although it also reported higher rates of device-related complications and atrial fibrillation. Four recent citation bursts indicate emerging directions in PFO closure research.

The most frequently cited articles in the field were also analyzed. Table 2 lists the top 10 most cited publications, each with more than 150 citations. The top three were all published in the NEJM. In 2012, Furlan et al. reported that device closure did not provide greater benefit than medical therapy for patients with cryptogenic stroke or transient ischemic attack (TIA) associated with PFO. In 2013, Meier et al. concluded that PFO closure for secondary prevention of cryptogenic embolism did not significantly reduce the risk of recurrence or mortality compared with medical therapy. In 2017, Mas et al. showed that PFO closure combined with antiplatelet therapy reduced the recurrence of cryptogenic stroke compared to antiplatelet therapy alone. However, several studies have also noted an increased risk of atrial fibrillation following PFO closure.

Table 3 lists the top 10 journals with the highest number of publications and citations in the field of PFO closure. The top three journals by publication volume are Catheterization and Cardiovascular Interventions (n = 110), Journal of Interventional Cardiology (n = 28), and Journal of the American College of Cardiology: Cardiovascular Interventions (JACC: Cardiovascular Interventions, n = 24). Among these, JACC: Cardiovascular Interventions has the highest impact factor (IF = 11.7), while Stroke has the highest average number of citations per article (42.38 citations).

Core journals in this field were identified using Bradford’s Law and visualized with Bibliometrix R (Figure 6). The results fully align with those presented in Table 3. Bradford’s Law, also known as the law of scattering, describes the distribution pattern of scientific literature across journals, highlighting both the concentration and dispersion of research output as well as its zonal structure (10). Among the top 10 journals by publication volume, only Catheterization and Cardiovascular Interventions has published more than 100 articles, while the publication volume of other journals is less than 30 articles.

Keywords encapsulate the core themes of scientific publications and reflect the focal points of research in a given field. Through analyses of high-frequency keywords keyword co-occurrence and clustering scholars can gain insights into the current status development trends and research hotspots in the field of PFO closure. Additionally keywords with strong citation bursts can help identify emerging frontiers in the domain. Using VOSviewer, 1,671 keywords were analyzed, and 126 high-frequency terms (occurrence ≥10) were identified. After merging synonyms and excluding irrelevant terms, clustering analysis revealed four major research themes (Figure 7A). The red cluster centers on PFO-related clinical mechanisms, featuring keywords such as paradoxical embolism, stroke, and right-to-left shunting, underscoring the link between PFO and cryptogenic stroke. Diagnostic advancements, such as dynamic imaging during echocardiography, reflect a shift toward functional risk assessment. The blue cluster focuses on diagnostic techniques, including transesophageal echocardiography (TEE) and contrast echocardiography, which remain the gold standards for PFO detection. Innovations like 3D-TEE and bubble studies now allow for detailed assessment of high-risk anatomical features, aiding clinical decision-making regarding closure. The green cluster highlights treatment approaches, with prominent terms including PFO closure, percutaneous transcatheter closure, and Amplatzer devices. Since 2015, pivotal trials such as RESPECT and CLOSE have influenced a shift in treatment strategies, favoring minimally invasive closure over medical therapy for high-risk patients. The yellow cluster addresses outcomes and epidemiology, with terms such as recurrence, prevention, and follow-up, emphasizing the efficacy of PFO closure in reducing stroke recurrence. Emerging evidence also suggests secondary benefits, including migraine relief, further expanding the therapeutic value of the procedure. Together, these clusters outline the evolution of PFO closure research, from understanding pathophysiology to advancing diagnostics and implementing evidence-based interventions.

The temporal overlay visualization generated by VOSviewer illustrates the dynamic evolution of research priorities in PFO-associated pathologies, with acute ischemic stroke emerging as a central investigative focus (Figure 7B). It reveals three distinct phases: from 2012 to 2014, research concentrated on pathophysiology (e.g., paradoxical embolism, atrial septal aneurysm [ASA]) and diagnostic validation (e.g., contrast echocardiography), establishing PFO as a modifier of stroke risk. The period from 2015 to 2016 marked a rise in interventional strategies, characterized by the adoption of percutaneous closure techniques and Amplatzer occluders, reflecting increased clinical interest following device-based trial outcomes. From 2017 to 2018, the focus shifted toward optimizing outcomes, with studies addressing residual shunts, thrombus resolution, and migraine comorbidity. This shift illustrates the maturation of PFO management into a multidisciplinary paradigm integrating neurology, cardiology, and imaging. Notably, the post-2016 emergence of headache relief as a prominent term underscores growing recognition of migraine associated with PFO as a distinct therapeutic target.

Burst detection analysis (Figure 7C) further captures evolving trends in PFO management. Early bursts, such as paradoxical embolism (2004–2011) and risk (2004–2008), reflect initial efforts to establish a link between PFO and cryptogenic stroke. A decline in interest in transcatheter closure (2007–2008) followed inconclusive early trials (e.g., CLOSURE I), while renewed attention in 2018–2019 and a sustained focus on PFO closure from 2020 to 2024 coincided with key trial results (e.g., RESPECT Extended Follow-up, REDUCE), which supported closure in high-risk patients (e.g., those with large shunts or ASA). At the same time, medical therapy (2015–2021) remained relevant for low-risk populations, reflecting ongoing debates around the use of antiplatelet agents. More recent bursts, such as antiplatelet therapy (2018–2022) and management (2020–2024), highlight a growing interest in personalized treatment strategies, incorporating risk stratification tools (e.g., RoPE score) and shared decision-making frameworks. The emergence of atrial fibrillation (2021–2024) indicates heightened awareness of the interaction between PFO and AF, prompting a shift toward more refined diagnostic protocols, including extended cardiac monitoring to exclude occult AF prior to closure. Collectively, these shifts mark a transition from empirical treatments to evidence-based, patient-tailored strategies, driven by technological advancements (e.g., improved occluders) and evolving clinical guidelines, such as the 2020 AHA/ASA recommendations.

Figure 7D presents a thematic map of research trends in PFO studies, generated using Bibliometrix R. This analysis evaluates themes along two axes: centrality (horizontal), which reflects a theme’s relevance within the broader research network, and density (vertical), which indicates its level of conceptual development. Motor Themes (upper right quadrant), including patent foramen ovale, risk factors, and PFO closure, represent mature areas of the field, supported by landmark trials (e.g., RESPECT, CLOSE, DEFENSE-PFO) that established the clinical benefits of closure in specific subgroups (e.g., patients with large shunts or ASA). Niche Themes (upper left quadrant), such as Amplatzer devices, reflect technological innovation, exemplified by the FDA’s 2016 approval of the Amplatzer PFO Occluder. Basic Themes (lower right quadrant), including atrial fibrillation and ischemic stroke, highlight ongoing diagnostic challenges, particularly in differentiating PFO-related strokes from those caused by AF. Emerging or Declining Themes (lower left quadrant), such as adult congenital heart disease, indicate shifting research priorities influenced by clinical trial criteria (e.g., DEFENSE-PFO’s exclusion criteria). This thematic framework illustrates the process of knowledge translation: Motor Themes align with guideline-recommended practices, while Niche Themes represent technological innovation, together advancing the field toward precision medicine in PFO management.

Research on PFO closure has undergone significant clinical evolution between 2004 and 2024. Early studies, such as those by Furlan et al. (2012) (11) and Meier et al. (2013) (12), reported inconclusive findings regarding the benefits of PFO closure over medical therapy for the prevention of cryptogenic stroke. These results initially dampened enthusiasm for closure as a treatment strategy. However, a landmark study by Mas et al. (2017) (13) published in the NEJM shifted the clinical paradigm by demonstrating that PFO closure combined with antiplatelet therapy significantly reduced the recurrence of cryptogenic stroke compared to antiplatelet therapy alone. This study reinvigorated interest in the field and contributed to a surge in research publications from 2018 to 2024.

Several pivotal clinical trials have helped establish the role of PFO closure in the secondary prevention of stroke. The RESPECT trial (2013) and its extended follow-up (2017) demonstrated that PFO closure significantly reduced the risk of recurrent stroke in patients with cryptogenic stroke. Similarly, the REDUCE trial (2017) confirmed the efficacy of closure in reducing stroke recurrence, particularly in high-risk patients. Collectively, these trials provide strong evidence supporting PFO closure as an effective intervention in carefully selected populations.

One particularly influential contribution to clinical practice has been the study by Kent et al. (14), which introduced the RoPE score. This tool helps differentiate between strokes that are likely attributable to a PFO and those in which the PFO may be incidental. The RoPE score aids clinicians in estimating the likelihood of stroke recurrence and in making informed decisions regarding PFO closure. Since its publication, this scoring system has been widely cited and has significantly influenced clinical guidelines and routine decision-making.

Our keyword burst analysis and thematic mapping highlight several ongoing clinical controversies that warrant further discussion: Upper Age Limit: Most randomized trials enrolled patients younger than 60 years; however, real-world registry data (Danish nationwide study (15)) suggest that PFO closure may also benefit older patients (aged 60–75) with high-risk anatomical or clinical features. Despite these findings, the optimal age threshold for closure remains a subject of debate (16). Residual Shunt: Residual shunts post-closure [observed in 5–25% of cases (17)] are associated with recurrent events (OR 3.1, 95% CI 1.4–6.9) (18). Advanced imaging (3D-TEE) and improved occluder designs (e.g., Amplatzer PFO Occluder) aim to mitigate this issue (19). TIA as a Qualifying Event: Current guidelines generally restrict PFO closure to patients with documented ischemic stroke, but emerging data [e.g., from the Gore REDUCE trial subgroup (20)] suggest potential benefit in high-risk TIA patients. Post-Closure Antiplatelet Duration: The 2020 AHA/ASA guidelines recommend 3–6 months of dual antiplatelets (21), yet recent studies propose individualized regimens based on thrombotic risk (22). Postprocedural AF Management: The Danish registry (15) reported a 7.8% 5-year AF risk post-closure, predominantly within 3 months. Early detection (e.g., with implantable loop recorders) and anticoagulation strategies are now prioritized (23).