Using Kano diagrams to display the most cited article types, affiliated countries, authors and MeSH terms on spinal surgery in recent 12 years

Background Citation analysis has been increasingly applied to assess the quantity and quality of scientific research in various fields worldwide. However, these analyses on spinal surgery do not provide visualization of results. This study aims (1) to evaluate the worldwide research citations and publications on spinal surgery and (2) to provide visual representations using Kano diagrams onto the research analysis for spinal surgeons and researchers. Methods Article abstracts published between 2007 and 2018 were downloaded from PubMed Central (PMC) in 5 journals, including Spine, European Spine Journal, The Spine Journal, Journal of Neurosurgery: Spine, and Journal of Spinal Disorders and Techniques. The article types, affiliated countries, authors, and Medical subject headings (MeSH terms) were analyzed by the number of article citations using x-index. Choropleth maps and Kano diagrams were applied to present these results. The trends of MeSH terms over the years were plotted and analyzed. Results A total of 18,808 publications were extracted from the PMC database, and 17,245 were affiliated to countries/areas. The 12-year impact factor for the five spine journals is 5.758. We observed that (1) the largest number of articles on spinal surgery was from North America (6417, 37.21%). Spine earns the highest x-index (= 82.96). Comparative Study has the highest x-index (= 66.74) among all article types. (2) The United States performed exceptionally in x-indexes (= 56.86 and 44.5) on both analyses done on the total 18,808 and the top 100 most cited articles, respectively. The most influential author whose x-index reaches 15.11 was Simon Dagenais from the US. (3) The most cited MeSH term with an x-index of 23.05 was surgery based on the top 100 most cited articles. The most cited article (PMID = 18164449) was written by Dagenais and his colleagues in 2008. The most productive author was Michael G. Fehlings, whose x-index and the author's impact factor are 13.57(= √(13.16*14)) and 9.86(= 331.57/33.64), respectively. Conclusions There was a rapidly increasing scientific productivity in the field of spinal surgery in the past 12 years. The US has extraordinary contributions to the publications. Furthermore, China and Japan have increasing numbers of publications on spinal surgery. This study with Kano diagrams provides an insight into the research for spinal surgeons and researchers.

research achievements (IRA) [4][5][6][7][8][9]. However, several changes have been encountered in those studies using bibliometric analyses, such as all factors (e.g., authors or affiliated countries/areas) in an article are viewed with equal contribution to an article. As such, the research achievements might be unfair and biased in IRA measurement. Two major difficulties we encountered in our study include the following: (1) coauthors with equal credits in an article byline are unreasonable; and (2) the Hirsch's h-index [7] has less discrimination power due to its integer nature (i.e., many of the same index value) making it difficult to differentiate the personal IRA [10]. The total citation count (e.g., author impact factor, AIF = citations/publications for each author) [4,5]) has the disadvantage of favoring researchers with few highly cited top publications or productive authors with many publications, but a relatively small number of citations [4,5].
While the h-index is determined by the maximum square that fits under the citation curve of an author when plotting the number of citations in decreasing order, the x-index [= max i (i × c i ) , where all the number of citations (denoted by ci) in the x core at publication [9] was proposed in 2018 and determined by the maximum rectangular area that fits under the curve. The disadvantage of the x-index is the equal importance placed on the citations and publications, regardless of the IRA tendency toward the citations or the publications. The Kano diagram [11] is thus suitable for interpreting the feature for factors in three parts: the excitement, the performance, and the basic requirement denoting the IRAs as the influential, the prolific, and the productive, respectively, in the previous articles [12,13].
Although many studies [14][15][16][17][18][19][20][21] have used bibliometric methods for evaluating worldwide research productivity in many biomedical fields, including those published in general orthopedics and its subspecialties [22][23][24], bibliometric analyses on the quantity and quality of articles published in spine journals worldwide are still rare [2]. Only three [1][2][3]  Therefore, the purpose of this study was (1) to evaluate the research citations and publications in major spine journals using the PMC database in recent 12 years and (2) to provide visual representations using the visual Kano diagrams onto the spinal research to display IRAs for surgeons and researchers.

Data source
We searched the PMC using "(( (((((((1529-9430)  All data used in this study were downloaded from the PMC, meaning that the study did not require ethical approval according to the regulations promulgated by the Taiwan Ministry of Health and Welfare.

Two prerequisites used for evaluating author IRAs
An author-weighted scheme (AWS) was applied to determine coauthors' contributions in article bylines [25] to improve the unreasonable phenomenon that all papers are equal weight irrespective of the number of coauthors [26,27]. Accordingly, more importance was given to the first (primary) authors and the last (corresponding/ supervisory) authors, while we assumed that the others (i.e., the middle authors) made smaller contributions [25].
Furthermore, the x-index [9] used in this study generalizes the h-index [7], which is determined by the square area (i.e., the number of publications equal to the citation point). The x-index thus surpasses the h-index in differentiating the personal IRA among affiliated countries/ areas, authors, or the medical subject headings (MeSH terms) [28].

Task 1: x-index for journals and article types
The trend of the number of articles across continents worldwide and journals over time was shown in a contingency table based on the 1st authors affiliated to their countries/areas. Journals and article types were compared to identify the most influential ones with either a higher x-index or h-index.

Task 2: x-index for affiliated countries/areas and authors
Three plots of the choropleth map [29], the bar chart, and the Kano model [11] were applied to display IRA features for affiliated countries/areas and authors based on the 18,808 abstracts and the top 100 most cited articles, respectively.

Task 3: x-index for MeSH terms
MeSH terms extracted from the top 100 most cited articles were represented by the topic burst on the topic of spine surgery [30]. Three plots of the Kano diagram, the line chart, and the clusters using social network analysis(SNA) [31] were used for interpreting (1) the evolution of topics in the past 12 years, (2) the most cited terms, and (3) the MeSH clusters are separated by the SNA using Pajek software [32]. The most cited article and the most productive authors in these 18,808 abstracts were extracted separately from the analysis performed on an author-made routine with codes of visual basic for application(VBA) on Microsoft Excel.

x-Index for journals and article types
A total of 18,808 publications were extracted, of those 17,245 were identified with affiliated countries/areas, and 46,795 authors were included. The 12-year impact factor for the five journals together is 5.758. The largest number of articles on spinal surgery was from North America (6417, 37.21%); see Table 1. The journal Spine earns the highest x-index (= 82.96); see Table 2. Comparative study has the highest x-index (= 66.74) among article types; see Table 3.

x-Index for affiliated countries/areas and authors
The United States has the highest x-indexes (= 56.86 and 44.5) among the world based on the total 18,808 publications and the top 100 most cited articles, respectively; see

x-Index for MeSH terms
A total of 82 MeSH terms were retrieved from the top 100 most cited articles. The most cited MeSH term was surgery with the x-index = 23.05. We can see the term of surgery with a red bubble on the right-bottom corner in the Kano diagram (top in Fig. 3), indicating the tendency toward its productive feature (i.e., with more weights on publications according to the definition of x-index).
The principal MeSH terms extracted from the top panel in Fig. 3 are present at the middle panel of Fig. 3, which shows that both costs of illness and disability evaluation occupy a lower proportion over the years. Both epidemiology and diagnostic imaging have accounted for more since 2016, implying the importance of these topics in spinal surgery in recent years. In contrast, the MeSH term of surgery has been dramatically decreasing since 2016, indicating a relatively smaller number of publications was on the topic of surgery.
The most cited article (PMID = 18,164,449) was written by Dagenais and his colleagues in 2008 [33]. The most productive author is Michael G.

Discussion
We observed that the largest number of articles on spinal surgery was from North America (6417, 37.21%) and Asia (5812, 33.7%). The results are different from the findings that the US and Europe are more dominant in science [34,35].
In the previous study [2], the largest number of publications in the field of spinal surgery was from the US (5137; 39.17%), followed by Japan (1408; 10.74%), and China (1131; 8.62%). The US had the highest h-index (106), followed by Canada (60), and the United Kingdom (54). However, our study observed a different result where the most productive countries are the US (5745, 32.95%), followed by China (1770,10.15%), and Japan (1712, 9.82%). The US also had the highest x-index (56.86), followed by China (39.06), and Japan (32.31); see the bottom panel in Fig. 1.
Furthermore, Spine earns the highest x-index(= 82.96) of all journals, and comparative study has an x-index (= 66.74) highest among article types. We have not seen any publication that applied x-index in bibliometric analyses to measure IRAs on spinal surgery in literature [1][2][3][14][15][16][17][18][19][20][21][22][23][24]. The x-index [9] generalizes the h-index [7], which is determined by the square area. The drawback of the x-index is the equal importance placed on the citations and publications that can be resolved by using the Kano diagram; see Figs. 1, 2, 3, which is a highlight of this study.   The second feature of this study is the AWS used for quantifying the contributions of authors (or their affiliated countries) in an article byline. A fair IRAs could not be achieved; otherwise, see Figs. 1 and 2. Similarly, the weights of MeSH terms can be quantified by the AWS with an equal size (e.g., Wi = citations/n, where n = the number of MeSH terms in an article and Wi is the weight of a MeSH term(i)) in an article and then plotted in Fig. 3.
Another feature is about the choropleth map incorporated in the bar chart and the Kano diagram as map legends designed for visualizing the metric disparities for better interpretation of values on the choropleth map [36]. For instance, the US with a yellow bubble is the most influential country in the performance area on the bottom panel in Fig. 1, providing a complemental interpretation of the choropleth map at the top panel in Fig. 1. All figures provide QR-codes for interested readers to examine the details on the information with the dashboard on Google Maps, different from those regarding spine-related studies [37,38] using the world map only available in color online.
We present the most cited MeSH term of surgery with the x-index = 23.05 based on the top 100 most cited articles. The method computing IRA for MeSH terms has never been seen before in academics although the placement scheme is similar to the coauthors suitable for quantifying their credits with the AWS in an article.
There are several limitations to this study. First, the five main journals relevant to spinal surgery in this study cannot represent other specialties and the many more basic research journals publishing articles related to spinal surgery. Nevertheless, these five major journals used in this study can represent the study and research contributed to the field.
Second, there might be some biases in author identification because some authors in the bibliometric database may have the same name or use the same abbreviations, but are affiliated to different institutions.
Third, the data extracted from the PMC cannot be generalized to other major citation databases such as the Scientific Citation Index (SCI; Thomson Reuters, New York, NY, the United States) and Scopus (Elsevier, Amsterdam, the Netherlands).
Lastly, many researchers believe that the impact factor or citation does not reflect the scientific quality of the research [39][40][41]. Many journals are now preferring citation distribution curve over the JIF [39]. It is worth discussing how this method for author and journal analyses can be included. Whether the x-index [9] that is applied in this study can be used for defining research achievements in a wider range of disciplines is worthy of further studies and discussions in the future (Fig. 4).

Conclusions
There was a rapid increase in scientific research productivity in the field of spinal surgery in the past 12 years. The US has exceptional contributions to