A bibliometric analysis of clinical research on fracture-related infection

Background Infection following orthopaedic trauma surgery is increasingly recognized as one of the major research priorities with as primary goal, improving patient care. This increased interest has been anecdotally recognized through published research, research grants, and, finally, with the development of the fracture-related infection (FRI) consensus group. In 2017, the accepted consensus definition of FRI was published, which has been followed by consensus recommendations from both a surgical and medical perspective. A bibliometric analysis was performed to objectively describe the trends in published clinical research related to FRI. Methods The terms related to FRI were searched in the Web of Science database between 2000 and 2020. The characteristics of clinical research on FRI regarding the author, country, journal, institution, scientific output, top 100 most cited articles, and trend topics were analyzed using Bibliometrix and WPS Office. Results A total of 2597 records were eligible for inclusion in this bibliometric approach, with studies originating from 89 countries, including eight languages. The United States of America (USA) published the highest number of articles and citations. International collaborations were present between 72 countries, with the most active country being the USA. The most contributive institution was the University of California. The highest number of papers and citations were from the Injury-International Journal of the Care of the Injured and the Journal of Orthopaedic Trauma. The top 100 most cited articles were published in 27 different journals, with the number of citations ranging between 97 and 1004. The latest trend topics were related to the diagnosis of FRI. Conclusion The present bibliometric analysis shows the research characteristics and trends of FRI from multiple perspectives. The fact that there is an increasing number of studies being published on FRI shows the agreement among scientists and clinicians that standardization with respect to this topic is very important.


Introduction
Internal fixation of fractures has revolutionized modern orthopaedic trauma care over the last 50 years. When indicated, internal fixation of fractures can preserve native joints and provide adequate stability to facilitate bone healing without limiting the functional rehabilitation of patients with prolonged immobilization [1]. However, postoperative infections complicate operative fracture management and occur in 1−11% of cases [2][3][4][5][6]. The diagnosis and treatment of postoperative infections are often complex. They require a multidisciplinary approach (i.e., surgeons, infectious disease physicians, and clinical pharmacists), often result in additional surgeries and a long duration of antibiotic therapy [7,8]. The health care system costs are immense. Trauma patients with a postoperative infection generate inpatient costs 6.5-8 times higher than their uninfected counterparts [9]. Despite adequate treatment, clinical and functional outcomes often remain poor, with high dependency rates and an inability to return to work [10].
The terminology surrounding postoperative infections has been historically inconsistent, which has hampered research efforts. Terms such as infection after fracture fixation, surgical site infection, and septic nonunion (or pseudoarthrosis) were often used interchangeably but without clear criteria to define the entity [11,12]. Therefore, in 2017, the term fracture-related infection (FRI) was defined by an international consensus group, including the AO Foundation, Orthopaedic Trauma Association, PRO IMPLANT Foundation, and European Bone and Joint Infection Society [13]. Over the course of the past decades, FRI has increasingly been recognized as a clinical research priority to improve the outcome of this sometimes devastating complication [14][15][16].
Bibliometrics is a method of statistical analysis used to assess a particular subject's characteristics and major developmental trends based on published research. Such an anal-ysis has provided useful insight into the global state and trends related to different medical fields [17][18][19][20][21]. Anecdotally, there has been a significant increase in interest in the topic of FRI in recent years through published papers, targeted research grants, and consensus statements. However, till today, bibliometric analysis related to this topic has not been performed. The present study was aimed at describing the overview of clinical studies and trends on FRI between 2000 and 2020.

Inclusion and Exclusion
Criteria. Inclusion criteria for further analysis were based on the following: (a) article describing a clinical study on FRI in orthopaedics; and (b) review article, meta-analysis, clinical trial, or guideline. Exclusion criteria were as follows: (a) book chapters, conference proceedings, editorials, errata, or letters; (b) basic science, animal, and cadaveric studies; and (c) nonorthopaedic related fracture or infection, such as of the skull or sternum.

Data Extraction and Bibliometric
Analysis. Data were identified and extracted by three authors individually (CL, AF, NH). All records were exported from the Web of Science database with text and excel document then imported into the software of Bibliometrix (University of Naples Federico II, Italy) and WPS Office (Kingsoft, China) to analyze the result, respectively [22]. Data analysis involves the author, article title, country, journal, h-index, institution, author keywords, language, number of citations, number of publications, and publication year. The impact factor and quartile of the journal were collected from the Journal Citation Reports 2020. The Bibliometrix software was utilized to construct data visualization of international collaborations, top 10 authors' average output, and trend topics.

Publication Output.
A total of 2597 articles met the inclusion criteria using the electronic Web of Science data-base ( Figure 1). The total number of citations related to FRI is 58,690 (51,109 excluding self-citations), with an average citation frequency of 22.6 times per item. Research papers were published in eight languages. More than 94% of articles were written in English (2458), followed by German (91), French (19), Czech (16), Serbian (4), Turkish (4), Spanish (4), and Italian (1). Figure 2 presents the specific amount of annual publications regarding FRI. Since 2008, the number of articles has exceeded 100 per year. The year 2019 ranked as the most productive year (253), followed by 2017 and 2018 (175 and 171, respectively).

Country.
For the United Kingdom (UK), publications from England, Northern Ireland, Scotland, and Wales were merged [23]. Finally, 89 countries contributed to the FRI publications ( Figure 3). Of these, the USA contributed the highest number of articles (814), followed by China (318), the UK (298), and Germany (266; Table 1). The top four countries with the most contribution, each presented more than 10 articles per year between 2016 and 2020 ( Figure 4).
Intercountry collaborations were found between 72 countries ( Figure 3). The USA had the largest number of collaborations with other countries (46), followed by the UK (35) and France (31; Table 2). The most frequent collaborations were between the USA and Canada (40), followed by the USA and Germany (39), then Switzerland and Germany (32; Table 3).

3.3.
Organizations. The top five most productive institutions are listed in Table 4. The most productive organization was the University of California system (48), followed by Harvard University (45) and Vanderbilt University (44). Of the top five most productive organizations, four were based in the USA,  3 BioMed Research International and the remaining was from the University Hospital Leuven, Belgium.

Authors.
A total of 9,842 authors contributed to FRIrelated studies. Metsemakers WJ has the highest number of publications with 32 articles, followed by Giannoudis PV (31) and Obremskey WT with 25 publications each ( Table 5). Figure 5 demonstrates the average output of the top 10 authors between 2000 and 2020.
3.5. Journals. All included articles were published in 339 different journals. Injury-International Journal of the Care of the Injured had the maximum number of papers (316), followed by the Journal of Orthopaedic Trauma (220) and International Orthopaedics (105; Table 6). The year of the first publication of an FRI study in the respective journal is depicted in Figure 6, with the majority appearing in 2008 and 2019 (24).  Data from the 2020 edition of journal citation reports showed 303 journals with an impact factor. The journal with the highest impact factor was The New England Journal of Medicine (74.699), followed by JAMA (Journal of the American Medical) Association (56.272), and Intensive Care Medicine (17.44; Table 7).      (Table 8). All included studies were published in 27 different journals, with most publications in the Journal of Orthopaedic Trauma (24), followed by Journal of Bone and Joint Surgery-American Volume (21), and Injury-International Journal of the Care of the Injured (9).
3.7. Trend Topics. Trend topics were identified by author keywords in Bibliometrix. Parameter values were set to a minimum number of keyword occurrences more than 20    (Figure 7). The top three most frequently used keywords were "infection," "fracture," and "complications." The earliest keywords were "external fixation," "children," and "humerus." The keywords "fracturerelated infection," "diagnosis," and "risk factor" were the top three latest emerging trend topics of FRI research.

Discussion
The current bibliometric study comprised a comprehensive analysis of the scientific output related to FRI, exploring characteristics from publications, languages, countries, institutions, authors, journals, most cited articles, and trend topics. In the current report, we provide scholars with essential references and suggestions for further investigation on FRI.
4.1. Global Publishing Trends. Although some research papers evaluated orthopaedic-related infection using biblio-metrics [24,25], to the best of our knowledge, this is the first article to fully assess clinical research on FRI. Our analysis shows that the number of related publications exhibited a significant increasing trend from 2000 to 2020. Of all languages, English was the main international scholarly language in this field. In addition, 65% of studies in languages other than English was in German. Several bibliometric analyses of medicine-related research also noted that German was the most common second language following English [19,26].   In 72 of 89 countries involved in international collaborations, the USA had the highest number of research collaborations with other countries. Most of the international collaborations were between North America and European countries. A greater international collaboration should be widely established in the future.

Organizations and Authors.
Among the institutions, four of the top five highest contributing countries originated from the USA. The University of California from the USA was the most productive institution globally in clinical studies focusing on FRI. Scholars from Europe and North America demonstrated a dominant position. The publications on the diagnosis and treatment principles of FRI also proved that most members of the FRI consensus group originated from Euro-American countries [27,28]. In addition, half of the top 10 most productive authors on FRI were involved in the FRI consensus group [13]. Metsemakers WJ, from the University Hospitals Leuven, contributed most to clinic research of FRI. Regarding the top 10 author productions over time, most authors showed increased stability and persistence for a longer duration between 2016 and 2020. The list of top 10 most productive authors in FRI may provide a valuable reference for future scientific conference invitations for FRI experts.

Journals.
Of the top 10 journals with the largest number of papers, Injury-International Journal of the Care of the Injured had the highest number of relevant publications, whereas Journal of Orthopaedic Trauma was the most academic influential journal on FRI. Compared with the bibliometric study, which excludes non-English literature [29,30], the present bibliometric research also provides more valuable journal information to non-English speaking countries. Hence, scholars will most likely benefit from journal information to further subscribe or track the most relevant jour-

Most Cited Documents.
The number of citations is a traditional indicator for the assessment of the value of a certain study. Our study displays the top 100 most impactful studies on FRI and provides a resource for clinical scientists. Of all publications, the most cited research paper was published by Govender et al. [31]. The authors found that an implant containing 1.50 mg/mL of recombinant human bone morphogenetic protein-2 in the treatment of open tibial fractures could reduce the incidence of postsurgical infections. The most cited review article was published by Trampuz and Zimmerli [32]. The authors summarized the pathogenesis, classification, diagnosis, and treatment of infections associated with fracture-fixation devices. 4.6. Trend Topics. Trend topics were identified from author keywords. In terms of the surgical site, scholars placed a greater interest in infection after humeral fractures, followed by the location of femur and tibia, with a current focus on ankle fractures [33][34][35][36][37][38][39]. Furthermore, research topics and concepts also altered over the course of the years. With the introduction of the FRI consensus group, the term "fracture-related infection" became the standard, with more recent topics primarily focusing on risk factors and diagnostic criteria of FRI [15,[40][41][42][43]. Prior to this, researchers focused on the treatment of FRI [44][45][46], with a focus on debridement, bone transfer, and antibiotic treatment. In addition, with the publication of the consensus definition of FRI, this concept appears to be gradually gaining ground in the scientific literature.

4.7.
Limitations. There are several limitations to the present study. First, only a single database was searched, with other databases and sources not included in the bibliometric analysis. Therefore, some potentially valuable information may have been missed [47]. Second, the FRI-related terms of the search strategy were based on the literature and personal     Third, conference proceedings were also removed from this bibliometric analysis, due to the potential of being published on two occasions, as a conference abstract and also as a full journal article [48,49]. Fourth, the number of citations is commonly used to assess the publication quality in bibliometric analysis. To discover high academic impact publications in the clinical research of FRI, we listed the top 100 most cited articles. However, self-citation, publication date, and controversial articles all likely had an impact on the number of citations [21].

Conclusions
The number of articles on FRI showed an increasing trend over the last 21 years. English was the primary language used for academic exchange on the topic of FRI, followed by German. In addition, the USA has the most number and impactful publications, as well as international collaborations. China has great development potential in this field. A broader international collaboration on FRI is required in the future. The most relevant and academic influential journals on FRI are the Injury-International Journal of the Care

Data Availability
Data is available from the first author upon request.