The Web of Science database (Clarivate) was chosen for the literature search, which we conducted on June 11, 2024, due to its high data quality and its recommendation by the creators of the bibliometric analysis software we used [12]. Our search used a prespecified algorithm to identify articles about experimental rat models of cardiac arrest. The keywords used included "heart arrest," "cardiac arrest," "return of spontaneous circulation," "ROSC," "cardiovascular arrest," "cardiopulmonary arrest," "advanced cardiac life support," "ACLS," "cardiopulmonary resuscitation," "rats," "rat," "cotton rat," and "cotton rats." These terms were combined using the Boolean operators “AND” and “OR” to optimize the search strategy. The exact search algorithm implemented is provided in Supplementary Material 1. The document types of interest were "Article" (original works), "Review Article," "Early Access," and "Proceeding Paper."

After the search, titles and abstracts were manually screened to filter out irrelevant articles, such as those focused on animals other than rats (including humans), those using isolated heart models, and those using models of cerebral ischemia that did not involve cardiac arrest. Finally, the bibliometric data for the top 100 most cited relevant articles were extracted in ".txt" format (Supplementary Material 2).

In this work, we focused on several key objectives designed to advance our understanding of experimental models of cardiac arrest in rats. Primarily, we sought to review and analyze the scope and characteristics of the most influential works on this topic to identify pivotal themes and gaps. This encompassed the following: (1) identifying the most influential contributors (journals, authors, institutions, and countries); (2) examining patterns of existing collaborations; and (3) analyzing the central themes and subject areas addressed within the articles.

The bibliometric analysis was conducted using the biblioshiny web application from the bibliometrix package in R ver. 4.2.3 (R Foundation for Statistical Computing) and RStudio ver. 2024.04.2+764 (RStudio Inc) [13]. To examine the structures of knowledge, we focused on the authors’ keywords. To improve the representation of thematic areas, certain keywords were automatically removed (e.g., terms used in the search strategy) or merged (e.g., synonyms) (Supplementary Materials 3, 4).

We manually reviewed the 100 most cited works to identify their specific topics of interest and categorize them in groups. The groups were the following: (1) neurology (studies focusing on the central and peripheral nervous systems); (2) cardiovascular (studies focusing on the structure and function of the cardiovascular system during and after cardiac arrest); (3) organ transplantation (studies investigating the effects of cardiac arrest on transplanted organs, such as the liver); (4) outcomes of cardiopulmonary resuscitation (CPR) in the context of local anesthetic toxicity; (5) cardiac arrest model description; (6) multiple organs; and (7) other topics.

The time periods used in the thematic analysis were determined automatically by the bibliometric software we used.

The database search yielded 1,872 results, with 1,597 remaining after applying filters for the document type. From these, the 100 most cited articles relevant to the topic of experimental cardiac arrest models in rats were further examined. The 10 most cited articles are presented in Table 1 [14–23], and all 100 articles are presented in Supplementary Table 1.

The 100 most cited works were published between 1980 and 2022 by a total of 416 authors. They were published in 44 different sources, and they had an average citation count of 106.7 per document.

To gain a more comprehensive understanding of the impact of these articles, we also calculated the citations per year (CPY) for the identified studies (Fig. 1, Supplementary Table 2). CPY was used as a complementary metric to adjust for the age of the articles and identify works that had a rapid influence within a short time frame. Studies from the 1980s and 1990s, such as those labeled 1, 3, and 6 in Fig. 1, have a high number of total citations (TC) due to their longevity [23–25]. However, their CPY values are relatively low, reflecting a decline in annual impact as newer research has emerged. These foundational studies, while still influential, are no longer at the forefront of current research discussions.

In contrast, more recent studies, particularly those published after 2010, such as 13, 18, and 20 in Fig. 1, show higher CPY values, indicating a rapid accumulation of influence within a relatively short timeframe [6,26,27]. These works are quickly becoming essential references in the field, highlighting the evolving research focus.

The most impactful studies, represented by black circles in Fig. 1, such as 21, 22, and 24, surpass the 75th percentile in both TC and CPY [14,15,17]. These works maintain both high cumulative influence and continued relevance, suggesting their lasting importance in the academic discourse.

Additionally, studies like 1, 10, and 6, represented by light grey circles, continue to show high TC but have lower CPY values, indicating they have had a more significant past impact rather than ongoing influence [23,25,28]. In contrast, dark grey circles, like 18 and 19, demonstrate rising importance with high CPY despite having fewer TC, suggesting these studies are still in the process of establishing their long-term impact [27,29].

Finally, studies represented by white circles with black outlines fall below the 75th percentile in both TC and CPY, indicating a more limited impact overall. These works, while still influential enough to be among the top 100, have not had the same degree of ongoing influence as others in the field.

The complete list of journals that published at least one of the 100 most cited works in experimental cardiac arrest research in rats is provided in Supplementary Table 3. Notably, three journals that fall within the core zone of source clustering according to the Bradford law, the Journal of Cerebral Blood Flow and Metabolism with an impact factor (IF) of 4.9 in 2023, Brain Research with an IF of 2.7, and Critical Care Medicine with an IF of 7.7, published a combined total of 35% of the top 100 articles. Furthermore, 16 of the 44 journals (36.36%), focus on neuroscience/neurology.

The top 10 most relevant authors based on the number of publications are presented in Supplementary Table 4, with each author contributing four to nine articles. The most relevant institutions listed as affiliations are the University of Pittsburgh, the Pennsylvania Commonwealth System of Higher Education, and Johns Hopkins University (Supplementary Table 5). Regarding the resident country of the contributing authors, the United States was by far the most prolific, followed by Japan and Germany (Fig. 2).

Fig. 3 displays the most frequently used publication keywords. A significant number of these keywords indicate a focus on neuroscience, including cerebral ischemia, neuroprotection, depolarization, and neuronal damage. The focus on neurology was also evident when the main topic of the included articles was extracted in a manual review: 72% in neurology, 7% in organ transplantation, 6% in cardiovascular, 4% in outcomes of CPR in the context of local anesthetic toxicity, 3% describing the cardiac arrest models themselves, 3% involving more than one organ, and 5% covering various other topics.

The co-occurrence network map we made to visualize the relationships among terms within the 100 most cited articles revealed five major thematic clusters (Fig. 4). Three of them concern neurology-related topics (cerebral ischemia and related outcomes, brain imaging metrics, blood brain barrier), one concerns the cardiovascular system, and one is about organ transplantation (liver).

The thematic evolution analysis covered the periods 1980–1996, 1997–2005, and 2006–2020 and demonstrates shifting research priorities and the emergence of new topics over time (Fig. 5). From 1980 to 1996, research predominantly focused on cerebral ischemia and cardiac emergencies, highlighting foundational concerns. From 1997 to 2005, the continued prominence of cerebral ischemia indicates sustained interest. New themes such as fibrillation, diffusion, and apparent diffusion coefficient also emerged. From 2006 to 2020, there was a continued focus on cerebral ischemia and fibrillation, and newer themes also emerged: hypothermia, nuclear magnetic resonance, and blood brain barrier.

In the collaboration analysis (Fig. 6), the United States emerged as the central hub in this network. Notably, the United States’s most frequent collaborations were with Japan (n=4), followed by China and Germany (n=3). Additionally, the United States collaborated twice with Korea and once each with Canada, Denmark, Greece, Italy, and Poland. Germany participated in a modest number of international collaborations, partnering once each with Canada, Poland, Switzerland, and Thailand. Poland, while less connected, collaborated twice with Serbia and once with Canada. Other notable, albeit infrequent, collaborations include those between the Netherlands and the Czech Republic and between Denmark and Sweden, each with a single instance of partnership. China and Greece also have one recorded collaboration.

This bibliometric study is the first on this topic to our knowledge and provides a detailed analysis of the 100 most cited articles on experimental cardiac arrest models in rats, highlighting seminal works and foundational methodologies. We found a predominant focus on neurology-related topics such as cerebral ischemia, diffusion imaging, and blood brain barrier integrity. Three important journals published a significant proportion of the articles. The geographic analysis shows the dominance of high-income countries, particularly the United States. Minimal between-country collaborations were found.

Two of the included studies were seminal works that were among the first to describe experimental cardiac arrest and resuscitation models in rats [30,31]. Interestingly, these studies outlined the two major methodological approaches to cardiac arrest induction that are still in use today: asphyxiation and electrical induction with endocardial stimulation. These models are valuable tools for further investigations into cardiac arrest and resuscitation techniques.

The first of them, a study by Hendrickx et al. [31] published in 1984 in Resuscitation, describes an asphyxial cardiac arrest model in rats used to study cardiopulmonary cerebral resuscitation. In that study, 78 anesthetized and ventilated rats underwent asphyxia by clamping the endotracheal tube for 5 to 15 minutes. Resuscitation, involving 100% oxygen ventilation, extrathoracic compressions, and intravenous adrenaline, was successful within 60 seconds in 65% of cases, with 50% surviving for up to 24 hours. The study assessed physiological parameters and neurological outcomes post-resuscitation. This model became a foundational reference for future experimental cardiac arrest studies and has influenced subsequent work by providing a reliable method for simulating ischemic injury and testing therapeutic interventions in a controlled setting.

The second study was published by von Planta et al. [30] in 1988 in the Journal of Applied Physiology and describes a standardized method of inducing ventricular fibrillation and CPR in anesthetized, mechanically ventilated rats. Ventricular fibrillation was induced by applying alternating current to the right ventricular endocardium, followed by chest compressions with a mechanical compressor. Successful resuscitation after 6 minutes of cardiac arrest was achieved in 8 of 14 rats. That study demonstrated the feasibility of its model for studying CPR and provided a reliable framework for exploring hemodynamic and metabolic changes during and after cardiac arrest. It significantly influenced future research by offering a controlled method for testing resuscitation strategies and assessing physiological responses in experimental settings.

As indicated by the keyword analysis and co-occurrence network, cerebral ischemia has consistently been the major topic of interest. Interestingly, experimental cardiac arrest models in rats have served a dual purpose; not only have they been used to elucidate the effects and outcomes of cardiac arrest [32], but they have also provided valuable insights into the effects of transient global cerebral ischemia [33]. This is confirmed by the largest thematic cluster in the co-occurrence network (cerebral ischemia and related outcomes), which contains research efforts on free radical formation, reperfusion injury and the use of antioxidants to ameliorate its effects, neurological outcomes and neuronal damage/cell death, the neuroprotective effect of hypothermia, glutamate toxicity, and reactive gliosis. Adjacent to that major cluster, the brain imaging metrics cluster contains terms related to diffusion imaging for the assessment of cerebral damage and recovery [18]. The third neurology-related cluster contains distinct research efforts on the integrity and physiological role of the blood brain barrier in the context of cardiac arrest–induced cerebral ischemia [34]. The cardiovascular cluster contains research efforts on ventricular fibrillation pathophysiology and therapeutic approaches [20]. Additionally, the organ transplantation cluster contains studies on liver transplantation and organ preservation, using cardiac arrest models to generate ischemic conditions [35,36]. This was not a finding we initially expected because organ transplantation is not a primary focus within the context of emergency medicine and cardiac arrest research. However, it is in line with recent efforts to increase the percentage of non-recovered cardiac arrest victims who become organ donors [37]. Some of the identified studies aimed to increase organ viability in non-heart-beating donors (NHBDs), thereby increasing the pool of organs available for transplantation. This research cluster also underscores the versatility of the rat cardiac arrest model by extending beyond its traditional use for studying cardiac function and neurological outcomes to address broader challenges in organ preservation and transplantation.

Regarding the evolution of prominent themes throughout the years, our analyses reveal shifting research priorities and the emergence of new topics over time. From 1980 to 1996, the primary focus was on cerebral ischemia. Between 1997 and 2005, a strong emphasis on cerebral ischemia remained, and significant research emerged on ventricular fibrillation and brain imaging metrics such as diffusion. In the most recent period, from 2006 to 2020, cerebral ischemia continued to be a central theme, and the prominent new research areas were hypothermia, nuclear magnetic resonance, and the effect of ischemic conditions on the blood brain barrier. Interestingly, the interest in hypothermia as a research focus was sparked by the publication of two seminal randomized clinical trials in humans in 2002. These highly influential trials demonstrated promising neurological outcomes in patients who received therapeutic hypothermia post-resuscitation [38,39]. Therefore, clinical research with a solid physiological and pathophysiological basis can spark significant interest and drive advances in experimental basic research.

Our analysis reveals a notable lack of highly influential works about experimental cardiac arrest models in rats that focus on organs other than the heart and brain. Most research has predominantly emphasized the neurological and cardiac aspects of cardiac arrest, and highly cited studies investigating the effects on other critical organs such as the kidneys, intestine, and lungs are absent. This disparity highlights a gap in understanding the systemic effects of cardiac arrest. Future research should address this imbalance by exploring the pathophysiological effects on and potential therapeutic interventions for these other vital organs. Such efforts would provide a more global view of cardiac arrest and could improve multiorgan protection [40] and recovery strategies.

The analysis of the most cited works reveals the significant role of specific journals in disseminating influential research on experimental cardiac arrest models in rats. Notably, three important journals collectively published 35% of the articles. This concentration of influential articles within these high-impact journals underscores their pivotal role in shaping the field and highlights the effectiveness of the Bradford law in identifying core journals within a specific research domain [51].

The Journal of Cerebral Blood Flow and Metabolism is particularly noteworthy for its focus on cerebral ischemia and related neurological outcomes, which aligns closely with the predominant themes identified in our keyword analysis. Similarly, Brain Research and Critical Care Medicine have significantly contributed to the dissemination of foundational research in both the neuroscience and critical care aspects of cardiac arrest.

Furthermore, the fact that 36.36% of the journals publishing the top 100 most cited articles focus on neurology both reflects the centrality of neurological outcomes in cardiac arrest research and highlights the collaborative nature of the field, in which insights from neuroscience are integral to understanding and improving resuscitation techniques and outcomes.

The analysis of author affiliations revealed that the United States holds the leading position in this field. Japan, Germany, Poland, and other high-income countries also show notable research activity. In addition, upper middle-income countries such as China, Serbia, and Thailand contribute to this research field, demonstrating growing interest and expanding research capabilities in these regions. However, as is the case with other research fields [52,53], lower income countries are absent from the list of contributors to highly influential articles, indicating a gap in global research participation and capacity. Bridging this gap could offer significant benefits because involving researchers from diverse geographic regions can introduce new perspectives, methodologies, and clinical contexts that might not be available in higher-income countries alone. By integrating a broader range of insights and experiences, the research community can enhance the generalizability and applicability of findings to more global populations, especially in terms of the varying health and disease patterns seen in different regions. Although scientific merit and the relevance of the research must always remain the primary considerations, fostering greater inclusion of underrepresented countries could contribute to more comprehensive advances in the field. Addressing this gap could also help to build research capacity in lower income countries, promoting more equitable global health solutions and collaborations.

We found minimal between-country collaboration, suggesting that although research efforts are widespread and international, there is still a significant opportunity to enhance global cooperation and interdisciplinary research partnerships. In preclinical experimental studies, particularly those involving specific animal models, collaborations—especially international ones—are relatively rare, perhaps because those studies are highly specialized and resource-intensive, often requiring strict standardization and consistency within individual research groups. It is important to exercise caution when interpreting our results about collaboration networks. In some cases, collaborations identified through a bibliometric analysis might not fully reflect meaningful scientific exchange and could be nominal or driven by factors such as funding requirements, rather than genuine scientific collaboration. As such, while expanding international partnerships holds great potential, these networks should be interpreted carefully to avoid overestimating the extent of scientific cooperation.

This study has some notable limitations, most of them inherent to bibliometric studies [54]. First, it was based on bibliographic data from the Web of Science database due to the constraints of the bibliometric software used (biblioshiny), which analyzes metadata from a single source. We chose Web of Science because it is recognized for its high-quality and comprehensive scientometric data. Second, a bibliometric analysis inherently lacks the depth of a thorough content review. To address this, we manually reviewed the articles to provide some additional context and insights into the methodologies used and topics addressed. However, an elaborate review of the methodologies and outcomes of each study extends beyond the scope and objectives of a bibliometric analysis. Future reviews on this topic could track the progression from small animal models, such as rats, to larger animal models and eventually human studies. This developmental process is important for understanding how findings from early-stage research translate into clinical applications. Small animal models are ideal for initial experimentation, but larger models are needed to better replicate human physiology and evaluate the efficacy and safety of novel interventions. By examining this progression, future reviews would provide a clearer framework for understanding how therapeutic approaches evolve from preclinical studies to clinical settings. Third, we chose to analyze the 100 most cited articles to focus on highly influential work on this topic. However, a high citation count does not necessarily indicate superior methodological quality or reporting clarity. Consequently, our analysis might include influential but methodologically weak studies and potentially overlook high-quality studies that have not yet gained significant attention in the form of citations. Therefore, readers should consider that our findings represent a snapshot in time. Ongoing updates to the literature are essential.