A Bibliometric Analysis of the Top 100 Cited Articles on Hepatic Magnetic Resonance Imaging

The purpose of this study is to guide the readers to the impact of the articles published on hepatic magnetic resonance imaging (MRI). We searched Scopus using 10 different search terms for hepatic MRI. The selected studies were thoroughly reviewed by two independent authors and any disagreement was sorted out by mutual consensus. The list of articles and journals was downloaded into an excel spreadsheet. Only the top 100 cited articles were selected by mutual consensus among all the authors. These articles were further read in the full-text form and were further categorized into subgroups. Three authors independently reviewed the top 100 selected articles, and subsequently data was extracted from them and analyzed. Our study showed that the highest number of top 100 cited articles on hepatic MRI were from Radiology (30 articles) followed by European Radiology (14 articles). The American Journal of Roentgenology, Radiographics, and Journal of Magnetic Resonance had seven articles each. The United States had the highest number of articles by region. Nineteen other journals contributed only one article each to the list of top 100 cited articles. The contribution of authors to the top 100 cited articles was reviewed; all the authors contributing with more than two articles to the highly cited articles are given in Table 3 in the supplementary material. The maximum number of articles were published during 2009 (14 articles), and for a five-year period, the maximum contribution was made during 2008-2013 (44 articles). Our analysis gives an insight on the frequency of citations of top articles on hepatic MRI, categorizes the subtopics, the timeline of the publications, and contributions from different geographic distributions.


Introduction And Background
Many academic institutions, private and public organizations are concerned about the quality and productivity of scholarly projects, as these are often the main parameters to prioritize resources. Bibliometric analysis is a powerful tool to determine the impact of a scholarly work on the scientific community [1]. The greater the number of the citations of an article, the greater is its contribution to the literature. Furthermore, by determining the citation frequency we can easily identify important issues and give suggestions for future research directions. However, there is a paucity of literature concerning top article citations on hepatic magnetic resonance imaging (MRI). The main purpose of this study is to guide the readers about the impact of the articles published on hepatic MRI.

Review
In January 2017, the Scopus library database was searched for the citations of published articles on hepatic MRI using the terms "hepatic MRI, liver MRI, hepatic magnetic resonance, liver magnetic resonance, hepatic imaging, liver imaging, hepatic radiological imaging, liver radiological imaging, hepatic magnetic resonance imaging, and liver magnetic resonance imaging." In this list, other imaging techniques such as computed tomography (CT) scan, endoscopic retrograde cholangio-pancreatography (ERCP), hepatobiliary iminodiacetic acid (HIDA) scan, and magnetic resonance cholangiopancreatography (MRCP) were not the focus of our search strategy and hence, all these articles along with other irrelevant articles were excluded. No language or time restrictions were placed on the search. Similarly, there was no restriction on the types of studies; however, studies discussing non-human subjects and case reports were excluded. The selected studies were thoroughly reviewed by two independent authors and any disagreement was sorted out by mutual consensus. The list of article and journals was downloaded into an excel spreadsheet. All articles and their corresponding journals were collected in a single search using the Scopus library. The articles were sorted on the basis of their citations using the option "Times cited" that provided us with a list of all selected articles ranked on the basis of their citation frequency. Only the top 100 cited articles were selected by mutual consensus among all the authors. These articles were further read in the full-text form and were further categorized into subgroups. Using the screening method proposed by Lim et al. [2], three authors independently reviewed the top 100 selected articles and the following data were extracted and analysed: journal name and its impact factor along with the number of citation and articles per journal, number of authors and his or her contribution along with his or her authorship position, publication year and frequency of topcited articles in five years range, country of the first author or country where the study was originally conducted, and type of article (review or original).
A total of 100 articles were selected on the basis of their citations, the number of citations ranged from 73 to 371 with a total number of citations equal to 13558 and a median number of citations equal to 467.52 (Table 1).  The total number of authors for 100 articles was 710 with a mean of 7.1 authors per article. The contribution of authors to the top 100 cited articles was reviewed and all authors contributing more than two articles to the highly cited articles are given in the supplementary table at the end ( Table 3) (Figure 2).

FIGURE 2: Articles according to the country of origin
Of all the 100 articles, 70 articles were original articles, 26 were review articles while only four were conference papers ( Figure 3).

FIGURE 3: Type of manuscripts for the top 100 cited hepatic magnetic resonance imaging (MRI) articles
In this extensive literature search, we found that the top two articles were about diffusionweighted MRI primarily used for focal hepatic lesion detection. Both these articles were cited more than 300 times and were published in Radiology. A study of 160 patients showed that history, examination, and tumor markers are not sufficient for the diagnosis of the focal lesions, indicating the importance of hepatic MRI in the detection and differentiation of focal lesions [3]. Hepatic MRI also helps differentiate benign lesions like hepatic adenoma, hemangioma, and focal hyperplasia from malignant lesions making it an important diagnostic modality [4]. In our analysis, 70% of all the top cited articles were about the applications and characteristics of hepatic MRI for the diagnosis of hepatic nodules and metastasis. Further analysis showed that of the 15 articles with citations greater than 200, nine were about the application of hepatic MRI for the diagnosis of focal lesions and the comparison of MRI with other imaging techniques in diagnosing liver lesions. This high citation indicates that there is significant interest in the application of hepatic MRI for the diagnosis of focal lesions and metastasis, and it may be the most frequent use of a hepatic MRI.
Overall, it was found that a significantly large amount of work on hepatology imaging focused on the applications of different contrast agents and the comparison of different imaging techniques along with its use for the diagnosis of focal or metastatic lesions. In the list of top 100 cited articles, only eight articles discussed the applications of hepatic MRI for hepatic fibrosis, arterioportal shunts, and complications of cirrhosis. While only one article in the top 100 cited articles discussed the uses of hepatic MRI for the diagnosis of hepatic abscess and to delineate hepatic blood flow, highlighting the need for further research in these fields. This can also have important implications for the journal editors and stakeholders in identifying and evaluating scientific literature in the field of hepatology and radiology.
It is also necessary to acknowledge the possibility of publication bias toward these articles and that hepatology and radiology literature, like other medicine-related fields, may be tied to pharmaceutical industries [5]. Nonetheless, understanding the important features inherent to the highly cited articles would facilitate other researchers and investigators to publish effectively. We believe that this paper will also help to highlight the subspecialties, subtopics, and applications of hepatic MRI that have not been given enough consideration. For instance, from our analysis, there has not been significant research on the use of hepatic MRI for diagnosing some infectious etiologies, particularly abscess causing organisms, especially in the developing countries. We feel that this issue could have greater future representation in the list. Also with time, as the disease burden of liver diseases increases, we anticipate an increase in the use of hepatic MRI in general as a diagnostic modality [6]. Furthermore, more than 98% of all the articles were published in the developed countries and only one article was from India. This signifies the possibility that the application of MRI for the diseases like hepatic abscess and cirrhosis might have been under-reported indicating that research directions could be redefined.
Our analysis showed that the most highly cited articles were published in 29 different journals. More than half (n=51) of all the 100 top cited articles were published in the Radiology or European Radiology and American Journal of Roentgenology-all these journals were dedicated solely to the imaging techniques but had an impact factor less than seven. Our findings are interestingly in contrast to the application of Bradford's law, a concept of bibliometric analysis suggested by Brookes [7][8]. Bradford's law states that most investigators obtain their citations from few journals which are famous in their field of expertise and the impact and citation frequency gets weakened when the study deviates from their core journals. Consequently, this tendency leads to a higher number of citations stemming from a few journals with high impact factor [7][8]. However, our study indicates that there might be other factors responsible for the number of citations an article gets. One possibility would be a growing trend of publishing articles in journals specified for the specialty in contrast to journals with a high impact factor. The Journal of Clinical Oncology and the American Chemical Society contributed only one article to the list in spite of very high impact factors. Moreover, in concordance with many other bibliometric analyses, most of the articles (n=39) were from the United States [9][10]. Germany, Japan, Italy, Korea, and Belgium combined contributed 45 articles to the list of top 100 articles, indicating that most of the work on hepatic MRI has been done in developed countries. India which has a huge load of a population suffering from fatal hepatic diseases where approximately one in every five Indians has hepatic disease had only one article in the list of top 100 articles.
Further scrutiny of the top articles showed that most (n=44) articles of the top cited articles on hepatic MRI were published from 2008 to 2013. These findings can be explained by the fact that hepatology is a dynamic field in which there has been an immense advancement and a huge contribution to the literature recently. Furthermore, it signifies that clinicians and researchers prefer to rely more on the latest studies. But readers should consider that some elapsed time is required for the articles to gain a considerable number of citations and to achieve significant coverage. That is the possible reason for the relatively small number of articles in our list published during 2014 to 2017. While the absence of any article before 1982 in our list of top 100 cited articles suggests the limited usefulness of old articles in the modern era, or that not significant research was done around that time as MRI was invented only in 1977 [11][12]. However, this trend could be due to a combination of many factors like the quality of an individual study, database limitation to track older articles, lack of internet resources in the pre-1990s, and the fact that original work is often published in hard copies like textbooks rather than on online journals [13][14]. Figure 1 in our study interestingly depicts a downward trend in the number of articles over the years.
As with other bibliometric analyses, our study also has some potential limitations. First, we used only online search on one database, omitting articles which might have been published in textbooks forms and this might have also resulted in missing the articles published before 1980 as Scopus has been established lately. However, to make our search strategy more sensitive, we also did search on Google Scholar to cross-check the number of citations and to include the missing articles. Second, the inherent problems associated with citation analyses, such as the bias linked to rely on the total number of times an article is cited, must be noted as well. Furthermore, bibliometric analysis traditionally favors older articles and often omits latest articles usually from the last ten years. However, in our study, more than 50% of the top 100 cited articles on hepatic MRI were after 2000. Lastly, we used only Scopus library for our search; it is possible that our top 100 cited articles might have a different number of citations if we had used any other database like Google Scholar as a significant difference between various databases does exist.

Conclusions
Overall, our study found that a large amount of work on hepatology imaging focused on the applications of different contrast agents and the comparison of different imaging techniques along with its use for the diagnosis of focal or metastatic lesions. Only one article in the top 100 cited articles discussed the uses of hepatic MRI for the diagnosis of hepatic abscess, and to delineate hepatic blood flow, highlighting the need for further research in these fields. Our study also indicates that most of the work on hepatic MRI has been done in developed countries and that most (n=44) articles of the top cited articles on hepatic MRI were published in the period between 2008 to 2013. Our analysis also showed that the most highly cited articles were published in 29 different journals with more than half (n=51) of all the 100 top cited articles published in the Radiology or European Radiology and American Journal of Roentgenology; all these journals were dedicated solely to the imaging techniques but had an impact factor less than seven. Our findings are in contrast to the application of Bradford's law which states that most investigators obtain their citations from a few journals which are famous in their field of expertise and the impact and citation frequency gets weakened when the study deviates from their core journals. However, our study indicates that there might be other factors responsible for the number of citations an article gets. One possibility would be a growing trend of publishing articles in journals specified for the specialty in contrast to journals with a high impact factor.