A bibliometric analysis of classic publications in web of science category of orthopedics

Background The past century has witnessed the rapid development of operation technique, surgical instruments, and knowledge of the diseases in orthopedics. In the academic history, a number of classic papers boosted the advancement for surgery. In this paper, we performed a bibliometric analysis, aiming to determine the most influential studies within the field. Methods Articles were searched from the publication year of 1900 to 2016 according to the Science Citation Index Expanded database of the Clarivate Analytics Web of Science Core Collection database. Two citation indicators TCyear and Cyear were employed to characterize the classic articles and the articles were identified and analyzed. Results A total of 30 classic articles with TC2016 ≥ 1000 in Web of Science category of orthopedics were identified, all written in English between 1961 and 2007 by nine countries. The minimal value of TC2016 was 1010; the maximum 3570; and the average 1591. Thirty classic articles were published in eight journals that were listed in the Web of Science category of orthopedics in 2016, and in two other orthopedics journals that were no longer tracked by Web of Science category of orthopedics as of 2016. Among the top 10 cited articles in both TC2016 and C2016, five articles barely received attention in the first few years after their publication, while they became cited more and more frequently in the last decade. Conclusion This study evaluated the development and trend of orthopedics research by adopting bibliometric analysis. It serves as a guide for investigators in the future research.


Background
The modern term orthopedics derives from the older word orthopaedia, title of a book published in 1741 by Nicholas Andry [1]. Two Greek words orthos and paedios serve as roots for orthopedic surgery. The former one means straight and free of deformity and the latter one means a child [2]. Orthopedic surgery demonstrates a rapid progress with several recent advances noted within orthopedic subspecialties [3][4][5], basic science [6], and clinical research [7]. Bibliometrics is a widely used tool to map the literature around a research field. It can help us to gain insight into the research focuses and future development of trends in orthopedic surgery. The citation number of a published article approximately reflects the popularity of the study and indicates the significance of the article in a certain field [8]. A thorough bibliometric analysis of classic articles helps investigators efficiently learns the history of developments and future directions of a research field. In this study, classic articles were identified and their characteristics were analyzed based on the bibliometric analysis method in the hope that it may guide investigators in this field.

Materials and methods
Our study was based on the Science Citation Index Expanded (SCI-EXPANDED) database of the Clarivate Analytics (formerly known as the Thomson Reuters and the Institute for Scientific Information) Web of Science (WOS) Core Collection database. According to Journal Citation Reports (JCR) of 2016 (InCites Journal Citation Reports dataset updated September 09, 2017), it indexes 8879 journals with citation references across 177 WOS categories in SCI-EXPANDED. In total, 302,299 documents (including 227,023 articles) were found in WOS category of orthopedics from the publication of 1900 to 2016 based on SCI-EXPANDED (updated on March 12, 2018). Two citation indicators TC year and C year were employed to characterize the classic articles. TC year is the total citation number from WOS Core Collection since publication to the end of the most recent year [9,10]. C year is the number of citations in the most recent year. C 2016 means the number of citation in 2016. TC year ≥ 1000 was used to retrieve the classic articles [11][12][13]. We inserted all the data for each article for each year into spreadsheet software, and manipulated them using Microsoft Excel2016 [14,15]. In addition, all hard copies of the 32 classic publications were found to check analysis information. Affiliations in England, Scotland, Northern Ireland, and Wales were reclassified as being from the United Kingdom (UK) [16].

Results and discussion
Document type and language of publication Analysis of document types and their citations per publication was earlier proposed [17]. A total of 32 classic publications (0.011% of 302,299 documents) with TC 2016 ≥ 1000 in WOS of orthopedics were found within two document types indexed in the WOS. Thirty classic publications were found to be document type of article including three of them belonging to both document types of article and proceedings paper. Two were published as document type of review. A review entitled "OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines" [18] was the only classic document published in the latest year of 2008 in orthopedics field with TC 2016 of 1394. Only articles were used for subsequent analysis because they included complete research ideas and results [19]. As a result, we identified 30 classic articles (0.013% of 227,023 articles) in the category of orthopedics, all of which were written in English. Such low percentage of classic publications can be also found, for example 0.048% and 0.063% of all documents in WOS categories of neurosciences [20] and psychology [12] respectively as well as 0.046% and 0.0049% of all articles in WOS categories of neurosciences [20] and surgery [11] respectively.

Publication years
In recent years, Ho's group proposed a relationship between total number of classic articles in a year (TP) and their citations per publication (CPP 2016 = TC 2016 / TP) by the decades in a WOS category as a unique indicator, for example WOS category of surgery [11], psychology [12], and neurosciences [20]. Thirty classic articles in WOS category of orthopedics were published between 1961 and 2007. The maximum value of TC 2016 was 3570, the minimum 1010, and the average 1591. Figure 1 shows the distribution of these 30 classic articles over the decades, and their citations per publication (CPP 2016 ). The 30 classic articles received a total of 47,735 citations. Only two classic articles were found in the decade of the 1960s, and no classic article was identified in the most recent decades. The 1980s was the most prolific period in terms of classic articles in orthopedics, which was different from WOS categories of the 1970s in surgery [11], the 1970s in psychology [12], and the 1990s in neurosciences [20]. Besides, the decade of the 1960s had the highest CPP 2016 of 2401. The earliest classic article in orthopedics field was "The etiology of chondromalacia patellae" [21]

Countries, institutions, and authors
There were 30 classic articles in WOS category of orthopedics by nine countries. Twenty-seven articles (90% of 30 articles) were completed in a single country from five countries and three (10%) were completed international-collaboratively from six countries. The USA took the first place by total classic articles with 18 (60% of 30 articles), followed by the UK (six articles; 20% of 30 articles), Sweden (three; 10%), Canada (two; 6.7%), and one for each of Australia, Brazil, France, Japan, and Switzerland respectively. The USA also published 16 of 27 single-country articles, two of three internationally collaborative articles, 16 of 30 first author articles, 14 of 28 corresponding articles, and four of five single-author articles.
In total, 18 (60% of 30 articles) articles were completed in a single institution from 17 institutions and 12 (40%) were completed inter-institutionalcollaboratively from 31 institutions. Only two institutions such as Case Western Reserve University in USA and Linköping University Hospital in Sweden published two classic articles in WOS category of orthopedics. Other 45 institutions had only one classic article. Linköping University Hospital was also the only one that published two single institution classic articles, first author articles, and corresponding author articles. Twenty-two of the 47 classic institutions were located in the USA followed by nine from the UK, five from Canada, three from Sweden, three from Australia, two from Japan, and one from Switzerland, France, and Brazil respectively.
Among the 91 classic authors of the 20 classic articles in WOS category of orthopedics, only A.I. Caplan from Case Western Reserve University in the USA and J. Lysholm from Linköping University Hospital in Sweden published two classic articles including one first author and one corresponding author articles. A.I. Caplan also Institute for Work and Health, Canada     (Table 2). Table 3 shows the 30 classic articles in WOS category of orthopedics with both citation numbers and rankings for TC 2016 and C 2016 . Total citations indicated high impact or visibility of an article in a research field. Due to the citations of publications in WOS Core Collection were updated weekly, the total citation number an article has since its publication to the end of 2016 (TC 2016 ) was utilized [9,10]. The advantage of TC 2016 is that they remain invariable and ensure repeatability compared with the index of citation from WOS Core Collection [12]. The history of a publication's citations with time has long been studied [51]. The citation history shows characteristics of the influence of an article after its publication. The highly cited articles would not always have high influence or visibility in research society [52]. Five of the top 10 articles (TC 2016 > 1800) still have a C 2013 ranked in the top 10. Figure 2 shows the citation history of classic articles that were ranked among the top 10 in both TC 2016 and C 2016 . Although some recently published articles within the past few years had great potential, they did not have a high TC 2016 . Thus indicator of C 2016 would be interesting to show high impact in 2016. A typical example is the article entitled "Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030" [22] which was the most impact classic article in 2016 with C 2016 of 411. A sharply increasing trend of citations can be found in this article. Similarly, the article entitled "Guidelines for the process of cross-cultural adaptation of self-report measures" had the same impact trend as the article by Beaton et al. [28] in the last decade. Other three articles including Harris et al. [23], Bohannon and Smith [24], and Insall et al. [25] had low citations after their publication and then had an increasing trend in the last 10 years. Classic articles by Fairbank and Pynsent [37], Ware [39], Ganz et al. [41], Sim and Wright [46], and Maher et al. [50] also had sharply increasing citations after publication. Table 4 reveals characteristic of highly cited and the most impact classic articles. The five classic articles were highlighted as follows:

Citation history of classic articles
Projections of primary and revision hip and knee arthroplasty in the USA from 2005 to 2030 [22] with C 2016 of 411 and TC 2016 of 2012.
Based on NIS, the study collected a substantially large number of discharge records, and revealed the information of the demand for primary and revision hip and knee arthroplasties in the USA through 2030 for the first time. It helped to quantify the expected number of hip and knee revision arthroplasties in the future. It also laid the necessary foundation for subsequent cost-benefit analysis nationally, to measure the increasing societal impact of revision arthroplasty in the USA.
Guidelines for the process of cross-cultural adaptation of self-report measures [28] with C 2016 of 363 and TC 2016 of 1870.
With the increasing number of multinational and multicultural research projects, there is a growing need to adapt the language of health status measures. The term "cross-cultural adaptation" is used to describe a process that involves both language (translation) and cultural adaptation issues in the process of preparing a questionnaire. This paper firstly presented a guideline for the process of cross-cultural adaptation of self-report measures, allowing equal efforts to collect data in crossnational studies and to avoid the selection bias.
Traumatic arthritis of hip after dislocation and acetabular fractures: treatment by mold arthroplasty: an end-result study using a new method of result evaluation [23] with C 2016 of 253 and TC 2016 of 3470.
The Harris Hip Score was initially introduced in this paper as a research tool to assess the clinical results of mold cup arthroplasty for traumatic hip arthritis. It made it possible for surgeons to compare their surgical outcomes in the literature. And it is the most widely Interrater reliability of a modified Ashworth scale of muscle spasticity [24] with C 2016 of 197 and TC 2016 of 2169.
The modified Ashworth scale is the most common clinical scale used to measure the increase of muscle tone and to monitor the course of disease. It was the first time that the concept of "Modified Ashworth Scale" had been proposed and that "grade 1+" had been added in the definitions. Meanwhile, the authors graded the elbow flexor muscle spasticity of 30 patients with intracranial lesions and proved the reliability of "modified Ashworth scale." Rationale of the knee society clinical rating system [25] with C 2016 of 161 and TC 2016 of 2161.
This paper presented a newly developed rating system for the knee. The knee society clinical rating system has been widely validated. The unified usage of it allows clinicians across the world to objectively compare their operational outcomes. Figure 3 shows trends of seven classic articles with sharp increasing in citations. These articles might be high impact in WOS category of orthopedics. In addition, classic author J.E. Ware also published the three classic articles about MOS 36-Item short-form [53][54][55].
Classic sleeping beauties in web of science category of orthopedics A "sleeping beauty" is a term that describes a research article that remains relatively uncited for a time and then suddenly bursts out. Van Raan [12,56] defined the three characteristics of "sleeping beauties" to be depth of sleep, length of sleep, and awakening intensity.   3. The intensity of the wakeup period: the number of citations per year for 4 years following the sleeping period Furthermore, long sleep and high impact sleeping beauties were also discussed [12]. Table 5 lists six high impact sleeping beauties in Web of Science category of orthopedics [12]. Figure 4 shows typical citation curves for four of them. The life of the article by Delee and Charnley [36] shown in Fig. 3 had the longest sleeping period with the deep sleep and the less deep sleep of 22 years respectively. The article by Outerbridge [21] was the earliest sleeping beauty while the article by Insall et al. [25] was the latest one in Web of Science category of orthopedics. Articles by Insall et al. [25] and Tegner and Lysholm [34] had higher impact in recent year. Furthermore, the article by Tegner and Lysholm kept in a plateau for 7 years after its sleep for 13 years and then "wake up" again to reach a high position in short period.

Conclusion
The bibliometric analysis provides a comprehensive overview of the most influential publications in the field of orthopedics. Based on our analysis, the decade with the most articles was the 1980s. All included articles belong to the document type of article and were written in English. The citation history of classic articles might serve as a guide to the understanding of the discipline.