Citation Classics and Trends in the Field of Opioids: A Bibliometric Analysis

Introduction Bibliometric analysis is one of the emerging and latest statistical study type used to examine and keep a systemic record of the research done on a particular topic of a certain field. A number of such bibliometric studies are conducted on various topics of the medical science but none existed on the vast topic of pharmacology - opioids. Hence, we present a bibliometric analysis of the ‘Citation Classics’ of opioids. Method The primary database chosen to extract the citation classics of opioids was Scopus. Top 100 citation classics were arranged according to the citation count and then analyzed. Results The top 100 citation classics were published between 1957 and 2013, among which seventy-two were published from 1977 to 1997. Among all nineteen countries that contributed to these citation classics, United States of America alone produced sixty-three classics. The top three journals of the list were multidisciplinary and contained 36 citation classics. Endogenous opioids were the most studied (n=35) class of opioids among the citation classes and the most studied subject was of the neurosciences. Conclusion The subject areas of neurology and analgesic aspects of opioids are well established and endogenous and synthetic opioids were the most studied classes of opioids. However, the egregious issues of addiction and misuse of opioids were underrepresented in the citation classics. The pulmonary and gastrointestinal aspects of opioids are also marginalized among the citation classics.


Introduction
Each year, extensive contributions are made to the medical science literature [1]. Bibliometric research provides quantitative analysis and extensive insight regarding the research conducted, the era in which the work is performed, and eminent authors, countries, and institutions; this approach also identifies aspects that remain to be examined [2], hence to monitor studies regarding a specific topic, practitioners refer to bibliometric researches. Although several bibliometric studies have been conducted in breast cancer [3], orthopedic surgery [4], epilepsy [5], thrombolytic therapy [6], and valvular heart diseases [7], our literature search revealed no such bibliometric analysis for opioids.
Opioids are the prototypical derivatives of opium, which is considered one of the world's oldest drugs [8]. Opioids are extensively used as analgesics, antitussive agents, antidiarrheal drugs, and anesthetic agents for management of acute pulmonary edema and various other treatments [9]. Some opioids, such as heroin, are illegal [10]. Moreover, the abuse or overuse of opioids can result in adverse effects and opioid dependence [11].
To fill this gap, we conducted a bibliometric analysis on opioids. The paper presents the opioid citation classics, that is, articles that were cited more than 400 times [12], as well as the trends in recently published articles.

Materials And Methods
A citation search was conducted to identify the 100 most cited articles and citation classics in the available literature (see Appendix). Like other researchers [6,7], we chose Elsevier's Scopus online database for our search, as Scopus provides 20% more coverage than Web of Science with more accurate citation counts than Google Scholar [13]. Full articles were accessed from PubMed, EMBASE (Excerpta Medica dataBASE), and Science Direct.
'Opioids', 'opiates', and 'opium' were our primary search keywords. To avoid bias and prevent missing relevant articles, we expanded our list to the names of individual opioids mentioned in Basic and Clinical Pharmacology [14]. The extended list included 'buprenorphine', 'butorphanol', 'codeine', 'diphenoxylate', 'fentanyl', 'heroin', 'levorphanol', 'loperamide' 'methadone', 'meperidine', 'morphine', 'naloxone', 'nalbuphine', 'nalmefene', 'naltrexone', 'oxycodone', 'propoxyphene', 'pentazocine', 'tramadol', and 'tapentadol'. All electronic database searches were performed on October 9th, 2017. Keywords were searched in 'article titles', 'abstracts', and 'keywords'. Relevant articles were retrieved and sorted by the option of 'Cited by', which yielded a list of articles arranged in descending order of their number of citations. No filters of language, time, human studies, subject area, territory, or affiliations were used. Abstracts and full texts of the articles were read from the sorted list and irrelevant articles were excluded from the analysis.
All article types, other than those requiring manual searching, telephone access, guidelines, and non-PubMed indexed articles, were included. The dataset was further evaluated by title, first and senior author, institution, department of the first author, topic, source, year of publication, and country of origin. In contrast with other researchers [6,7], we used CiteScore [15], Source Normalized Impact per Paper (SNIP), and SCImago Journal Rank (SJR) to rank the journals. Some articles were cited more frequently than others, due to differences in the time since publication. We adjusted for this error by determining citation index for each article.
Citation analysis of the extracted articles was conducted both by using Scopus and by manually screening the articles. Articles were further classified into two broad categories: 1) major subject area and 2) class of opioids. Tables and charts were created using Microsoft Excel 2016 (Microsoft Corporation, Washington, US). IBM Statistical Package for the Social Sciences (IBM SPSS Statistics for Windows, Version 20.0. Armonk, New York, US) was used to apply the Pearson product moment correlation co-efficient to evaluate the relationship between citation times, CiteScore, and citation density. Chi-square test was applied to observe significant difference between the categorical values and Non-parametric tests, such as the Mann-Whitney U test and Kruskal-Wallis test, were applied to evaluate any significant difference between categorical and numerical data. P-value < 0.05 was considered significant in all cases.

Trends of citation classics
Top citation classics were published from 1957 to 2013. Most were original articles (n=80), while the remainder were review articles (n=17), conference papers (n=1), and letters to the editor (n=2). The citations of those articles summed to 92,413, ranging from 545 to 3,149, with a median of 762.5 and a mean of 924. 13. Approximately 4% were self-cited and 3.6% were book citations, reducing the original citations to 85,482. Citation density (citations per year) ranged from 10.53 to 147.63, with a median of 33.06 and a mean of 38.97. A moderately positive correlation was found between citation count and citation density (r=0.511), while no significant correlation was found between citation count and year of publication. There was no significant difference between the citations of original articles and other article types (p= 0.311). Figure 1 shows the overall trend of total citations by year, with the peak citations in 1988.

Origins, institutions and authorships
The citation classics were produced by nineteen different countries. Figure 2 shows major contributors to the top 100 citation classics, with the leading contribution from the United States (n= 58). Five hundred authors contributed to the citation classics, with a median of four authors and mean of five authors. The number of authors per article ranged from 1 to 21. Details of authors with more than two citation classics are shown in Table 1

Journals
The citation classics were published in 40 journals. The CiteScore of these journals ranged from 2.45 to 29.6, with a median of 8.67 and a mode of 7.87. More than half of the articles (n= 62) were published in nine journals ( Table 2). Weak positive correlations were found between CiteScores of the journals and the number of articles published (r=0.255), total citations of the journals (r=0.253), citation count of the articles (r=0.080), citation density (r=0.085), and years of publication (r=0.086). There was no significant difference between CiteScore by subject area (p=0.432), classification of opioids (p=0.076), or type of article (p=0.267).

Subject areas and topical distribution
Citation classics belonged to four major subject areas: neurosciences, anesthesiology and pain management, gastroenterology, and general medicine. Figure 3 shows the number of articles published each year by specific subject area. The distribution of citation count was same in all subject areas (p=0.260). No significant difference was found between subject areas and citation density (p=0.136) and year of publication (p=0.105). Forty-six of the citation classics primarily discussed opioids' mechanism of action, followed by adverse effects (n=24), clinical uses (n=13), pharmacokinetics (n=4), basic sciences (n=3), and opioid tolerance (n=2). The remaining seven discussed various other topics.

Classes of opioids
Citation classics were distributed among four major opioid classes and a miscellaneous topics category. Thirty-five articles discussed endogenous opioids, followed by synthetic opioids (n=30), opium (n=8), and semi-synthetic opioids (n=4). The remaining 23 citation classics discussed opioids in general. Significant differences were observed between classes of opioids and years of publication (p=0.037), citation count (p=0.000), and citation density (p=0.000). Figure 4 shows the number of articles on classification of opioids published in each subject area. There was no significant difference between articles published on different classes of opioids and subject areas (p=0.491).

Discussion
The use of Opium dates back to the commencement of the Greek era [16]; opioids are opium derivatives [17]. The oldest opioid articles available on PubMed were published in 1945. The fact that none of these pioneer articles is a citation classic suggests that the quality of the work plays a greater role in citation number than does the number of years for which the article has been a part of the literature [12].

Trends of citation classics
Citation classics ranged over 53 years of publication, with zero to seven articles per year. Seventy-two of the citation classics were published from 1977 to 1997. The most-often cited article in the list was published in 1988. In contrast with previous studies [7], our graph of the total article citations over time (Figure 1) showed several peaks and troughs. After 1995, a gradual decrease in citations occurred, followed by a rapid decrease in 1997. Only 24 of all articles published after 1997 were citation classics. This finding supports the idea that some topics undergo intense study at certain times during which extensive research is performed, and after which the topic ceases to be of broad and current interest [18]. There was no significant difference between the citations of original research articles and review articles (p=0.310), contradicting the belief that review articles are more often cited [7]. Interestingly, only two articles had a citation density greater than one hundred, and the article ranked seventeenth had the highest citation density of 147.63 by Chou R et al.

Origins, institutions and authorships
Sixty-three of the citation classics came from the United States (Figure 1). Campbell explains this major contribution from the USA by stating that reviewers and authors from the USA show bias towards local papers [19]. The United Kingdom produced the second-highest number of citation classics (n=13), followed by France (n=10). Thirteen of the citation classics were produced through the contributions of researchers in multiple countries.
In total, 160 institutions contributed to our citation classics. Forty institutions had more than two articles among the citation classics. Thirty-five of the citation classics had multiinstitutional origins. Of these, 13 papers were also multinational, suggesting that international collaborations produce high-quality output that greatly benefits the scientific community [7].
A total of 500 authors contributed to the citation classics. Each of the top 12 authors contributed to at least three articles. Bloom F, although 12th in rank, had the highest H-index, followed by Watson SJ, Kosterlitz HW, and Akil H. Authors with a high H-index not only have a greater chance of having their work accepted, but are also more likely to get promotions and become reviewers [20].  Table 2). Physiological Reviews, a medicine-based journal, had the highest CiteScore (29.60) overall, but only one citation classic was published in it.

Journals
According to the Bradford law, our top three multidisciplinary journals that had the highest citations are Zone 1 or core journals [21]. Similarly, the journals belonging to the categories of neurosciences and medicine are Zone 2 and Zone 3 journals, respectively. This trend indicates that although opioids are part of the field of pharmacology, citation classics on opioids were primarily published in multidisciplinary and neurology-based journals, which contradicts an established concept that highly cited articles are published in field-specific journals [13]. We used multiple analytical parameters to rank our journals, including CiteScore, SJR, and SNIP, to reduce bias [22]. CiteScore is a metric similar to a journal's Impact Factor that gives us a comprehensive view of the journal's effect on the scientific community.

Subject area and topical distribution
Neurosciences seemed to be the most studied subject area, as half of the citation classics covered topics in neurology, psychology, cognitive sciences, and psychiatry. This finding is similar to previous reports [23]. The most likely explanation is that the mechanism of action, clinical uses, tolerance, and addiction involve the neural system. Thirty-one of the citation classics addressed anaesthesiology and pain management, demonstrating the major role of opioids in this subject area. Even though opioids have major effects on the gastrointestinal tract [24], we observed that gastroenterology articles were underrepresented among the opioid citation classics. The work in various subject areas was uniformly distributed among years of publication.
As mentioned above, citation classics primarily discussed the opioid mechanism of action, covering the structure, saturation, availability, binding sites, and location of opioid receptors in various organs. Although deaths due to opioid overuse and addiction continue to skyrocket [25], 24 articles focused on the adverse effects of opioids. In these articles, drug dependence, abuse, and addiction were highlighted. Only one article among the citation classics examined prescription protocols.

Classes of opioids
Opioids are classified as endogenous, opium alkaloids, synthetic opioids, and semi-synthetic opioids [16]. Endogenous opioids, which include endorphins, enkephalins, and dynorphins, were most-studied. Twenty of the 35 citation classics regarding endogenous opioids fell under the subject area of neurology ( Figure 4). The second most cited class was synthetic opioids. Thirteen of these 30 citation classics fell under the subject area of neurology.
Some limitations must be considered. Possible citation bias, including in-house citations, negative citations, and incomplete citations is a major limitation. In addition, the use of one database to extract the list may have overlooked articles that were not recognized by Scopus, and excluded textbooks. Scopus has been reported to miss older citations, which results in omissions of research conducted and published before 1980 [26,27]. Our list may have missed some citation classics; this can be explained as 'obliteration by incorporation' [12], which in simple terms states that the content of some classic articles has become such common knowledge that they no longer require citation.

Conclusions
We conclude that the neurologic and analgesic aspects of opioids are well established in the top 100 citation classics. Endogenous and synthetic opioids were the most studied. Although addiction and misuse of opioids is a worldwide menace to health and economy, this concern was underrepresented in the citation classics. We would also like to draw attention to the pulmonary and gastrointestinal aspects of opioids, which are marginalized among the citation classics. Similarly, quality papers on opium alkaloids, semi-synthetic opioids, and substance abuse are needed.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. 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.