Characteristics and Trends of the Most Cited Publications in The Journal of Arthroplasty

Background This study aims to identify the most frequently cited articles published in the Journal of Arthroplasty (JOA) and to analyze the trends in the content and contributors of the literature within the journal. Methods The 100 most cited articles published in the JOA were accessed using the Scopus database. The number of citations, year of publication, level of evidence (LOE), article type, country of origin, and contributing institution were each recorded for each article. Results The United States (63%) was the most prolific publishing nation. The 1990s (30%) and 2000s (47%) were the most productive decades. The most common article category was clinical outcomes (33%), followed by technical note (16%) and biomechanics (14%). The plurality of the top 100 articles were well-designed case-control or cohort studies of LOE II (46%) followed by LOE V (32%) and LOE I (11%). Conclusions Using citation analysis, the most influential articles in the JOA were comprehensively and objectively analyzed. The most popular fields of research involved clinical outcomes (33%) and technical note (16%), both of which increase an article’s likelihood of being highly cited. Knowledge of the most influential articles in the JOA allows for appreciation of current and potential future areas of literature regarding diagnosis, management, and outcome of a patient undergoing arthroplasty.


Introduction
Peer-reviewed literature remains the foundation of how evidence-based medicine is practiced by clinicians all around the world. Total joint arthroplasty (TJA), including both total hip arthroplasty (THA) and total knee arthroplasty, comprises a large cohort of surgical procedures performed in the United States of America (USA) [1]. The ever-growing body of literature across numerous journals regarding arthroplasty provides a challenge to sift through and find the highest quality of evidence studies that can directly impact patient care.
Bibliometric analysis has become a popular method to identify and analyze specific topics or trends by using the most influential articles in each topic or journal. The analysis is run by order of citations to determine the impact and influence of each article. This method has been used to highlight topics of interest, shed light on under-researched topics, and determine the quality of contributions from various orthopaedic journals [2,3]. This method has also been useful for analyzing orthopaedic injuries and procedures such as meniscal injuries, unicondylar knee arthroplasty, and hip and knee arthroplasty [4e6].
The Journal of Arthroplasty (JOA) is one of the top journals in orthopaedic surgery and the number 1 journal focusing on joint arthroplasty of the hip and knee when measured by impact factor. There is, however, lack of a coherent summary of the most relevant content of this journal. Our study aims to analyze the characteristics and trends of the top 100 most cited articles in the JOA. The purpose of this analysis is to elucidate the influence this journal has had in its various areas of research focus. We hypothesize that most of the literature comes from research groups out of the USA and are focused on clinical outcomes regarding prosthetic joint infection. Table 1 The top 100 most cited articles. 7 Muratoglu

Material and methods
In September 2021, the Scopus database was used to identify articles published in the JOA. Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were used in the collection of data. We began by defining the topic of this study as a bibliometric analysis of articles published in the JOA. Our search term in the Scopus database was "Arthroplasty," and our "search within" was set to "source title." We searched the journal since its inception in 1986 for all studies it has published. The articles were sorted in descending order of times cited. All included studies were written in English and peer reviewed.  A combination of Scopus metrics and data collection via author review was performed using Microsoft Excel. The following data were extracted from each article: title, study design, main topic, citations, year, institutional affiliation, country of origin, and level of evidence (LOE). The countries of origin were determined by the locations of the authors' affiliated institutions. The order of the top 100 articles was determined by the number of citations per each article. If multiple articles had the same number of citations, then the most recently published article had a prioritized ranking. The LOE was either provided upon evaluation of the abstract or determined by full-text review using the Oxford LOE Guidelines. The lead author (DL) determined article classifications of each study which included the following: surgical technique, clinical outcomes, anatomy/biomechanical, natural history, clinical guidelines, classification, imaging, technical note, and biomechanics. The following topics were used to classify publications: primary hip arthroplasty, primary knee arthroplasty, prosthetic joint infection, hip and knee arthroplasty, revision hip arthroplasty, primary shoulder arthroplasty, healthy knee, venous thromboembolism, implant, and revision knee arthroplasty. The topic of TJA was applied to articles that evaluated joint arthroplasty in more than 2 joints. Articles were placed into only one category via reviewer determination, and this single most applicable category for that article was selected. These data were verified by a separate investigator, with discrepancies being reviewed by the lead author.

Results
The top 100 most cited articles meeting our inclusion criteria were analyzed. All articles were published between 1987 and 2018. The greatest number of citations in 1 article was 914, while the least number was 169 ( Table 1). The total amount of citations was 26,278, which provides an average of 263 citations per article. In terms of productivity by decade, the 2000s were the most productive (n ¼ 47 publications), followed by the 1990s (n ¼ 30 publications), the 2010s (n ¼ 18 publications), and lastly the 1980s (n ¼ 5 publications) (Fig. 1).
Fourteen countries contributed to the top 100 articles. The USA had the greatest number of contributions with 63 articles. The United Kingdom (UK), Canada, and Australia followed with 8, 7, and 5 contributions, respectively. All other countries had less than 4 contributions (Fig. 2).
The most recurring LOE was II (n ¼ 46 articles) and V (n ¼ 32 articles). The remaining number of publications at each LOE was as follows: I (n ¼ 11 articles), III (n ¼ 5 articles), and IV (n ¼ 6 articles) (Fig. 3).
Articles were classified into 9 different article types. The most frequent article type was clinical outcomes (n ¼ 33 articles), followed by technical note (n ¼ 16 articles) and biomechanics (n ¼ 14 articles). All other articles types have 10 or less publications (Fig. 4). Articles were classified into 11 different topics. The most frequent article topic was primary hip arthroplasty (n ¼ 33 articles), followed by primary knee arthroplasty (n ¼ 30 articles), TJA (n ¼ 13 articles), and prosthetic joint infection (n ¼ 7 articles) (Fig. 5).

Discussion
This bibliometric analysis highlighted the most influential articles published by the JOA. The 100 articles were comprehensively analyzed across categories of publication, LOE, country of origin, authors, and frequency across decade.
Since its establishment in 1991, the American Association of Hip and Knee Surgeons has brought the most up-to-date information on a wide array of scientific topics to its members. Through its original, peer-reviewed articles, the JOA has been on the forefront of publishing high-impact medical literature. Previous bibliometric analyses have highlighted the impact that the contributions from this journal have had in the most influential THA and unicompartmental knee arthroplasty research [7,8]. A recent study analyzing authorship trends in the JOA found that over time there has been a significant increase in last authors with an MD/PhD and MD/MBA, increases in contribution from international authors, and increases in the mean number of authors per article [9]. The present study highlights the most influential articles published by this journal that have further developed the advancements in arthroplasty research.
Notably, 57% of the publications had an LOE of I or II which is slightly higher relative to other orthopaedic journals. In 2005, Obremskey et al. evaluated the LOE in various orthopaedic journals and found 32% of articles to have an LOE of I or II [10]. In the current study, 11% of the articles had level I evidence while 46% had level II evidence. Over the past 20 years, there has been a significant increase in the number of articles with LOE of I and II possibly due to the increasing emphasis on publication quality in orthopaedics [11].
Additionally, through categorization by article type, we were able to analyze patterns in the top 100 most cited arthroplasty articles. Specifically, clinical outcomes of various interventions comprised about one-third of this list (33%), followed by studies of technical note (16%) and biomechanical studies (14%). Other similar studies have found that clinical outcomes of arthroplasty dominate the orthopaedic literature with respect to citation frequency [3,8].
The most cited article was published by Kurtz et al. in 2012, and it emphasized the financial implications of periprosthetic joint infections. The research team projected that with an increasing demand for joint arthroplasty, cost of infected revisions to US hospitals may exceed $1.62 billion by 2020 [12]. The JOA recently published an article with similar evidence pointing toward increasing cost projects primarily due to increases in the total number of these procedures being performed [13]. The third most cited article also came from the work of Dr. Kurtz. This article   focused on infection burden for hip and knee arthroplasty [14]. It was published in 2008, during the decade which has had the greatest number of publications on our list. This further exemplifies the academic growth in arthroplasty research during this time span.
The second most cited article followed 147 patients undergoing acetabular component revision and classified their acetabular defects. Their paper stated that by adhering to the used classification system and utilizing the appropriate surgical technique, acceptable and predictable results of acetabular revision can be expected [15]. While most publications in this analysis focused on primary hip or knee arthroplasty, there has been increasing recent research in revision arthroplasty. Specifically with revision THA, there has been a focus on clinical outcomes with most papers having an LOE of II [16]. However, in this analysis, the authors did not find trends changing over time.
Furthermore, the USA is the country of origin for the majority of the articles in this list (63%), followed by the UK (8%) and Canada (7%). This follows the major trend in bibliometric studies, with America contributing the most to medical journals, especially in orthopaedic journals. However, a 2013 study evaluating knee arthroplasty and soft-tissue surgery reported that the USA had declined in publishing over the past 16 years, while the UK and Japan became more prolific in publishing [17]. Nonetheless, the USA continues to contribute the most to top 100 lists in terms of citation frequency [8]. We can attribute this to most renowned medical journals originating in the US, publishing in the English language, and more funding opportunities [18].
There were several limitations of this study. A bibliometric analysis that uses total citation count to rank publications naturally presents bias toward older articles because there has been a longer time period for these articles to accrue cumulative citations. Article classification and LOE were assigned by author review, which involves subjective interpretation despite the use of standard LOE guidelines. The country of origin analysis may not account for the potential multinational collaboration. Outside of Scopus indexing and updates reporting, it cannot be independently verified that all of the most recent JOA articles are indexed within the Scopus database.

Conclusions
Using citation analysis, the most influential articles in the JOA were comprehensively and objectively analyzed. The most popular fields of research involved clinical outcomes (33%) and technical note (16%), both of which increase an article's likelihood of being highly cited. Knowledge of the most influential articles in the JOA allows for appreciation of current and potential future areas of literature regarding diagnosis, management, and outcome of a patient undergoing arthroplasty.

Conflict of interest
Michele D'Apuzzo is a paid consultant at Zimmer Biomet and is a board member of the Florida Orthopedic Society and American Academy of Orthopedic Surgery; all other authors declare no potential conflicts of interest.

Informed patient consent
The author(s) confirm that informed consent has been obtained from the involved patient(s) or if appropriate from the parent,   guardian, power of attorney of the involved patient(s); and, they have given approval for this information to be published in this article.