Analysis of the Most Frequently Cited Articles in Hand and Wrist Surgery: A Modern Reading List

The aim of this investigation was to analyze the 50 most frequently cited articles on hand and wrist surgery of all time and those published during the 21st century. We aimed to evaluate the article, author, and journal characteristics for these publications in order to create a modern reading list of impactful upper-extremity articles. A search of the Journal Citation Reports 2022 edition to identify journals containing possible hand or wrist-related articles was performed. Related journals were identified and then searched on the Web of Science database to identify hand and wrist articles. The top 50 most cited articles overall and the top 50 most cited articles from 2000-2021 were identified and indexed. Several bibliometric parameters, such as study type, study topic, study design, level of evidence, citation count, citation density, the institution of the lead author, the gender of lead and senior authors, and country were analyzed. For the most cited articles of all time, the number of citations ranged from 224 to 1109 with a mean of 368 citations and 15.0 citation density. Citations for the top 50 articles from 2000-2021 ranged from 153 to 950 with a mean of 233 citations and 14.5 citation density. For both groups, the most common level of evidence was level IV (33% and 27%). No correlation between journal impact factor and citation count or citation density was found. In both groups, “fracture” was the most common topic and papers were predominantly written by male authors. Frequently cited publications on hand and wrist surgery are often clinical papers that contain low levels of evidence and tend to focus on topics related to fracture care. Female authors remain underrepresented.


Introduction And Background
There has been a substantial increase in the volume of peer-reviewed academic articles published during the past few decades across a variety of medical subspecialties [1]. Recent investigations have indicated that there were over 8,000 orthopedic surgery articles published between 2010 and 2014 from US-based institutions [2]. As article volume has continued to increase, so too has the number of orthopedic surgery journals. A recent publication reported that there were around 300 indexed orthopedic surgery journals as of 2020 under the Scimago database [3]. As both journal and article volumes continue to increase, upperextremity surgeons may face challenges in navigating the available literature to identify impactful and relevant articles [3][4][5][6]. "Classic" papers have helped build the foundation of modern practices by describing and analyzing the basic science concepts, techniques, outcomes, and complications of commonly performed procedures. While a variety of metrics are available to quantify an individual paper's influence, the most common and convenient way continues to be the number of citations the scientific work has received [7]. In this context, bibliometric analyses of top-cited articles can aid in understanding the characteristics of classic papers.
The purpose of this investigation was to identify and analyze the 50 most frequently cited articles on hand and wrist surgery. We aimed to evaluate the article, author, and journal characteristics for these frequently cited publications. In addition to identifying the most cited articles of all time, we aimed to define the most frequently cited hand and wrist articles published in the 21 st century in order to create a modern reading list of impactful upper-extremity articles.

Search Criteria
In February 2022, we conducted a search of the Journal Citation Reports 2022 edition to identify journals 1 1 1 1 1 containing possible hand or wrist-related articles. The Journal Citation Reports, published annually by Clarivate Analytics, is a database of basic science and social science academic journals [8]. This database provides tools to identify, categorize, and compare journals based on journal topics and citation metrics. We identified journals under the categories, "ORTHOPEDICS", "SPORTS SCIENCES", and "SURGERY" and refined the search with the terms "hand OR wrist". Journals were then searched on the Web of Science database to identify hand and wrist articles. The Web of Science is a Clarivate Analytics database that continually indexes and updates information on scholarly articles [8]. The platform allows users to search articles by journal or topic and track an article's citations or cited references. The top 1000 most cited articles were exported from the Web of Science database and then sorted by "times cited" in descending order. The articles were then reviewed by two authors for inclusion criteria (whether the identified articles were pertinent to orthopedic hand and wrist surgery) from the most to least cited until 50 articles were included. Articles were excluded if they did not focus on the surgery of the hand or wrist.

Data Extraction
From the remaining articles, the top 50 most cited articles overall and the top 50 most cited articles from 2000-2021 were determined. For all included articles, data from the Web of Science database for the following items were included: title, lead author, senior author, journal name, number of citations, citation density, and publication year. Citation density is determined by dividing the total number of citations by the number of years the article has been published. Additionally, two reviewers reviewed articles and recorded the study type (clinical, basic science), study topic (fracture, osteoarthritis, outcomes, etc.), study design, level of evidence, the institution of the lead author, gender of the lead and senior authors, and country. For the purposes of our investigation, we used the last name in the author list of each article to designate senior author status. None of the authors of this analysis were authors of the articles that were indexed in our investigation. The level of evidence was assigned according to the Journal of Bone & Joint Surgery guidelines (see Appendices) [9]. Where the gender of the author was undiscernible, the Genderize.io API was used to determine gender by first name [10]. Any disagreements in the classification of articles were reconciled by a third reviewer. Any disagreements in the classification of articles were reconciled by a third reviewer. Overall, there were only five disagreements where the initial reviewers could not settle on the level of evidence of an article. The 2020 two-year impact factor was obtained from each journal's respective web page.

Statistical Analysis
Descriptive statistics were measured to summarize the results. Means and percentages were used to compare articles from all years and articles from 2000-2022. A chi-squared test of independence analysis was used to compare author gender between groups. P-values < 0.05 were considered significant. All data analyses were conducted using IBM SPSS Statistics Version 28.0.2.2(15) (IBM Corp., Armonk, NY).

Results
The top 50 most cited hand and wrist articles overall and from the 21 st century are compiled in Tables  1-2 Table 3 includes the article, author, and journal characteristics. The fracture was the most common article topic (n=12), as noted in Figure 1. The most common level of evidence was level IV (33%), with only four studies (11%) containing level I evidence. All publications were written in English and 78 were published by US institutions. Of the 36 institutions that produced articles within the top 50, the Mayo Clinic published the most (n=5), followed by SUNY Upstate (3) and Roosevelt Hospital (3). Table 4 contains article citations per decade. The 1980s produced the highest volume of publications (14) and had the highest mean citations per decade (428) ( Table 5). The Journal of Hand Surgery (American volume) published over half (56%) of the most cited articles. The journal impact factor was not found to correlate with the number of citations (r: 0.003) or citation density (r: 0.115). The majority of authors were male with 82% of first authors and 92% of senior authors.        Table 2. Articles were published between 2000 and 2016. Citations ranged from 153 to 950 with a mean of 233 citations and 14.5 citation density. The most frequently discussed topic was fracture (22%), as noted in Table 4. Of clinical articles, the most common level of evidence was level IV (27%). All articles were written and English and originated from 15 countries and 12 journals. The Journal of Hand Surgery (American volume) published the most articles (44%). The impact factor of journals did not correlate with the number of citations (r: 0.129) or citation density (r: 0.145). There was a total of 42 institutions that published articles and the University Innsbruck (n=3), as well as the Washington University (n=3), published the most. Similar to the overall most cited articles, the gender of first authors (76%) and senior authors (89%) was mostly male.

Discussion
In aiming to determine the content domains of the most frequently cited articles in hand and wrist surgery, we found that articles pertaining to fracture management were cited most frequently. Fracture remained the most common article designation for papers published since 2000, indicating little change over time. Previous bibliometric analyses have been performed within various orthopedic subspecialties and have assessed content domains [11][12][13][14][15][16][17][18][19][20][21][22][23][24]. Similar to our findings, Huo et al. noted that fracture was the most common topic appearing within the 50 most cited articles in elbow surgery [23]. Kelly et al. found a prosthetic joint replacement to be the most common topic in their study analyzing the top 100 cited orthopedic surgery papers [13]. Within orthopedic trauma, distal radius fractures were reported to be the second most commonly discussed topic after hip fractures [16]. Considering the subspeciality overlap relative to fracture care in the upper extremity, it remains likely that fracture articles will continue to have a large impact with respect to citations.
With respect to the overall level of evidence contained within the most-cited article lists in hand and wrist surgery, our results indicated that level IV studies appeared most frequently. Prior studies looking at older temporal ranges of hand and wrist literature have similarly demonstrated that the most common study type is that of level IV evidence [12,18,19]. Previous studies have reported on the increase in level I evidence studies within orthopedic surgery over the recent years [12,18,19,25]. Despite the increase in impactful randomized, controlled studies on hand surgery, these papers remain far less cited than retrospective case series. This trend seems to be echoed in other fields of medicine too: A previous analysis showed that 36% of the top-cited oral surgery papers consisted of low-level evidence studies [26]. Similarly, a bibliometric analysis of 100 top-cited ophthalmology papers showed the most common level of evidence to be Level III [27]. This study has a number of limitations that should be considered. First, this study measures the impact of journals and ranks articles based on the number of citations. Citation-based metrics used exclusively do not directly measure the quality of literature. To truly assess the quality of a journal or study it is necessary to read the work, which is not always practical in studies assessing large quantities of work such as this. Lowlevel evidence papers receiving a high number of citations are common in other orthopedic subspecialties as well [7,28,29]. Goedderz et al. reported that 38% of the papers in their analysis of top-cited calcaneus fracture papers were of level-IV evidence [28]. Similarly, Tang et al. reported that level IV evidence studies made up 37% of the papers in the 100 top-cited anterior cruciate ligament reconstruction papers [29].
Considering the increase in level I studies over the recent years, it remains possible that papers with higher levels of evidence may permeate the frequently cited article lists.
We found that 18% of the overall top-cited and 24% of the top-cited papers since 2000 had female first authors. Previous bibliometric analyses of hand and wrist surgery publications have not reported on the gender demographics of authors, so it is difficult to compare them to historical data [7,18]. Historically, orthopedic surgery has been shown to have the lowest percentage of women and to recruit women at lower rates when compared to other surgical specialties despite an increasing number of women entering medical school [30,31]. A recent analysis of authorship in hand surgery, however, indicates an increase in the number of female authors in hand surgery research and that there has been substantial progress toward improving gender diversity in academic hand surgery over the last 14 years [31]. Despite this promising trend toward more even representation, the current bibliometric analysis found that the majority of first authors and senior authors continued to be male. As women are increasingly represented within orthopedic groups, future investigations should endeavor to assess increases in female representation among authors.
Of interest, we found no correlation between the journal impact factor and the number of citations a study receives for the most frequently published articles. Our results highlight some of the limitations surrounding the use of journal impact factors as a measure of "prestige" among authors and readers. Additional factors, such as publication immediacy or lag, ease of access to journals, self-citations in editor letters, and the use of non-source items, could also affect the impact factor of any given journal while not necessarily increasing the quality of the scientific work that is being published in it [32][33][34][35][36][37]. While a journal's impact factor might be a useful tool to gauge its qualitative properties, authors and readers should not use it to assess the quality of individual articles and they should be aware of its inherent limitations [33,36]. Other factors outside of the inherent articles' qualities may contribute to an increased number of citations. In orthopedics and other medical fields, studies with "significant" results (ie. studies that show statistically significant results with α<0.05) are cited twice as frequently as those without significant results [38]. Self-citation, "in-group" citation, and the authority of the senior authors may also contribute to citation bias in some cases [39,40]. Additionally, the increase in social media uses to promote scholarly research may necessitate a reevaluation of how we measure article impact [41].
This study has a number of limitations that should be considered. While a journal's impact factor might be a useful tool to gauge its qualitative properties, authors and readers should not use it to assess the quality of individual articles and they should be aware of its inherent limitations. Impact factors measure the average possible citations per study within the journal, however, the impact factor is highly influenced by a small proportion of articles with high relative citations [33]. Therefore, a majority of these articles are not highly cited. Also, we used categories and search terms to filter the body of current literature to articles pertaining to hand and wrist surgery. This system has the potential to miss articles relating to hand and wrist surgery simply due to how articles are cataloged. Furthermore, the citations analyzed within this study were those within academic journal articles and thus do not capture any recognition articles may have received in other media such as news outlets, textbooks, podcasts, or other web-based sources. Lastly, female authorship has the potential to have been underestimated by this study, as only the first and senior authors were analyzed. Given the small sample size, this investigation was likely insufficiently powered for any formal statistical comparisons between groups.

Conclusions
This study provides a detailed account and bibliometric analysis of the 50 most cited hand and wrist surgery publications of all time as well as the 50 most cited articles published in the twenty-first century. Frequently cited publications often contain low levels of evidence and tend to focus on topics related to fracture care. Female authors remain underrepresented. As the number of peer-reviewed journals and articles continues to increase, these data may function as a concise, modern reading list of impactful publications for hand and upper-extremity surgeons.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.