Bibliometric Research in Neurosurgery: A Review of the 50 Most Cited Bibliometric Publications in the Neurosurgical Literature

Bibliometry is a popular research method that is used to explore and analyze large volumes of data in an effort to highlight trends, patterns, and impacts within a specific field. This review aimed at highlighting the characteristics and citation patterns of the high-impact bibliometric research studies that were published in the neurosurgical literature. Using PubMed and Google Scholar, the 50 (52 due to identical citation numbers for the lowest three articles) most cited bibliometric research publications were identified and reviewed. Information relating to the articles’ publication and bibliometric features were retrieved. The articles’ citation numbers were collected. The median article age and journal impact factor (IF) were eight years and 2.76, respectively. The majority of studies were published in World Neurosurgery and the Journal of Neurosurgery, which were the publishing journals for 18 (35%) and 12 (23%) articles, respectively. Twenty-six (50%) articles were first authored by researchers from the United States of America (USA). The highest bibliometric component was science mapping, which was the theme in 30 (58%) articles. The majority of the bibliometric focus was clinical topics/fields (22 (42%) articles) and neurosurgeons/departments (21 (40%) articles). The most popular bibliometric metric was the h-index (±variants), which was employed in 22 (42%) articles. The median size of analyzed data was 188, and the most frequently utilized databases were Scopus (22 (42%) articles) and Web of Science (21 (40%) articles). The median (range) citation numbers were 52 (29-238). The citation analysis showed significantly higher citation numbers for older articles (aged ≥ 8 years) and studies published in the Journal of Neurosurgery. The citation rates were not influenced by the size of the data, the searched databases, or the bibliometric features. In conclusion, the most cited bibliometric research publications in the neurosurgical literature were predominantly descriptive analyses of clinical topics/fields and performance analyses of neurosurgeons/departments. Their citation numbers were relatively modest and were positively influenced by the publication’s age and by a specific publishing journal but not by the bibliometric features of the study. Bibliometric research provides useful analytic tools that can be utilized in review studies and other practical purposes such as scholarly practices and policy decision-making.


Introduction And Background
Bibliometric research encompasses a set of validated statistical methods that are used to analyze the literature in order to explore trends, patterns, and impacts within a specific field.Compared to a systematic review that summarizes and combines the findings of the existing literature on a specific research topic, bibliometric analysis sums up large quantities of data to describe the state of intellectual structure and emerging trends of a study field [1,2].Bibliometric practices have proved valuable across a wide range of disciplines, including medicine, science and technology, social sciences, education, and business and management [1][2][3].Researchers use bibliometric analysis for a variety of reasons, such as detecting changing shifts in journal performance, collaboration patterns, and research elements [3].The real value of bibliometric assessment remains in its capacity to process, categorize, explore, and report complex data.Additionally, it has the ability to present a network of ideas and topics in meaningful ways that enable researchers to identify knowledge gaps, derive novel concepts for investigation, and place their intended contributions to the field [1][2][3].
Bibliometric tools are quantitative by nature; however, they can be used to make statements about qualitative features.In fact, it has been suggested that the main purpose of bibliometric reviews is to transform something unquantifiable (scientific quality) into an assessable entity [4].Bibliometric techniques can easily be scaled from micro (institute) to macro (world), and the evaluation of research can be carried out at the level of the journal, researcher, department, medical specialty, country, and worldwide regions [4,5].The fundamental components of bibliometric evaluation are performance analysis and science mapping [1,2].Performance analysis focuses on the appraisal of the output of research in a given field.It involves the use of publication-and citation-related metrics for the assessment of research sources (articles, journals), domains (subject fields), and contributors (authors, institutions, countries) [1,2].Science mapping concentrates on the intellectual interactions and structural connections among research constituents.It makes it feasible to uncover the key matters along with the salient trends and gaps while shedding light on new developments in the field [1,2].
Citation analysis is an important basic technique in science mapping that functions on the assumption that citations reflect intellectual links between publications that develop when one publication cites the other [2,6].Citation count might not be a criterion for quality assessment; nevertheless, articles with higher citation numbers are considered a milestone in any field and can affect the research and clinical approach.Furthermore, it is recognized that an article's citation number will affect the publishing journal's impact factor (IF) and can be regarded as reflective of the article's endorsement, efficacy, quality, and the author's reputation [6].Citation analysis allows researchers to identify the most cited publications in their field.Assessment of the most influential publications in any subject will enhance knowledge of research evolution and highlight subjects of relevance in that area.Evaluation of high-impact studies in specialties, subspecialties, journals, clinical topics, and research types has been a matter of interest that received attention in recent years [7][8][9][10][11][12].Lately, bibliometric evaluation of systematic reviews and metanalyses has been a focus of several publications [9,11,12].However, bibliometric assessment of bibliometric studies remains a topic that is limited to a few reports in the literature [6,13].The purpose of this review is to identify and analyze the most cited bibliometric research studies that were published in neurosurgical literature.The study aimed to highlight the characteristics of bibliometric studies in the field of neurosurgery and to determine the factors that affect the citations among the 50 most influential articles on the subject.

Search Strategy
This study was carried out at King Saud Bin Abdulaziz University for Health Science, Jeddah, Kingdom of Saudi Arabia.No ethical approval was necessary by our institution as the study was based on data obtained from open-access sources.The PubMed database was searched on 15th December 2023 for suitable articles using the following combinations: (Title) Bibliometric OR Bibliometrics OR Cited OR Citation OR Citations OR Productivity OR Output OR Index OR Indices OR Level of Evidence OR Rank OR Ranking OR Rankings AND (Journal) Individual by name.The list of neurosurgical and spine journals searched and the number of screened articles are shown in Table 1.

TABLE 1: List of the searched neurosurgical and spine journals
A bibliometric publication was defined as being one of the following [1,2]: (1) studies in which quantitative techniques were applied to bibliographic data, such as publications and citation metrics; (2) studies that evaluated performance whether at the levels of individuals, institutions, countries, subject areas, or journals; and (3) studies that analyzed the most influential publications and those that looked at trends, relationships among citing publications, topics, authors [1,2].The search yielded a total of 1,348 studies, of which 1,071 were excluded due to being duplicates, non-bibliometric, or not providing adequate data.
Using Google Scholar, the citation numbers for the remaining 277 articles were documented.In view of the regular changes in the citation numbers, the findings on a single day (30th March 2024) were recorded and used for analysis.The 50 most cited articles (52 because of identical citation numbers for the lower three articles) were identified and chosen for this review.The selection was limited to bibliometric research studies published in the neurosurgical and spine journals, which will be referred to hereinafter as neurosurgical journals or neurosurgical literature in this article.A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing the flow of the review phases is presented in Figure 1.

Analysis of Articles' Characteristics
Using the full articles, information relating to each of the selected studies was collected by two of the authors independently, and any discrepancies were resolved by consensus.The following data was collected: Article publication features: publishing year, journal and its IF, number of authors, number of centers, number of specialties, number of countries, number of references, and the first authors' countries.Articles bibliometric features: bibliometric component (performance analysis, science mapping), bibliometric focus (neurosurgeons/departments, journals, clinical topics/ fields), bibliometric metrics (h-index (± variants), most influential studies, publication trends), data source (neurosurgical, medical journals), searched databases, analyzed data size, and the reporting of at least one significant finding.Missing data were referred to as not available (NA).The journals' IF data were obtained from an online source [14].

Results
The 52 most cited bibliometric research studies in the neurosurgical literature are summarized in

Articles Bibliometric Features
The distribution of the articles according to their bibliometric components, focuses, and metrics is illustrated in Figures 3-5.The number of articles based on the bibliometric component was science mapping: 30 (58%) and performance analysis: 22 (42%).The number of articles based on the bibliometric focus was clinical topics/fields: 22 (42%), neurosurgeons/departments: 21 (40%), and neurosurgical journals: nine (17%).The number of articles based on the bibliometric metric was h-index (± variants): 22 (42%), most influential studies: 15 (29%), and publication trends: 15 (29%).The number of articles according to data sources was medical journals: 44 (85%) and neurosurgical journals: eight (15%).The median (range) analyzed data size was 188 .The most commonly utilized databases and number of articles were Scopus: 22 (42%), Web of Science: 21 (40%), Google Scholar: nine (17%), PubMed: five (10%), Journal Citation Report: two (4%), and others: six (12%).A report of at least one significant finding was found in 30 (58%) articles.The median (range) article citation numbers were 52 .Tables 3-4 summarise the correlation and secondary analysis findings between the citation numbers and the various articles' characteristics.The correlation analysis showed no significant association between citation numbers and any of the publication and bibliometric features.The secondary analysis, however, demonstrated significantly higher mean citation numbers amongst older articles (aged ≥ 8 years) (P=0.0392) and in studies published in the Journal of Neurosurgery (P=0.0085).None of the other parameters reached significance.

Discussion
Reputable journals and researchers have published highly cited articles utilizing bibliometric methods to explore the progress and emerging trends in various specialties.For bibliometric research to be useful and contribute to advancing theory and practice, it is essential to be of high quality [1,2].Mukherjee et al. [1] identified seven factors that are useful in developing and evaluating effective bibliometric research.These are the following: novelty (what's new?), value (so what?), importance (who cares?), timeliness (why now?), exposition (why so?), rigor (well done?), and completeness (done well?).Furthermore, for the reporting of bibliometric research to be judged sound, several parameters should be mentioned clearly in the article.These include clear objectives, comprehensive systematic search using multiple databases with well-defined inclusion and exclusion criteria, suitable use of bibliometric indicators, proper checking of data for accuracy and normalization, appropriate data analysis utilizing advanced techniques and software tools, clear and informative visualization, and reasonable contextualization of findings [1][2][3].
The median citation number for the most cited bibliometric research studies published in the neurosurgical literature was 52 citations.This was lower than the citation numbers for higher levels of evidence research studies such as the top 100 trials on glioblastoma multiforme (median 349 citations) [7].It was also lower than the citation numbers for the top 50 survey research publications in the neurosurgical literature (median 111 citations) [8].Variation in citation rates according to study design and subject is well recognized in the literature [69].We found that the age of the publication (≥ 8 years) was a significant predictor of citation numbers.We also observed a positive link between citation rates and the bibliometric study being published in the Journal of Neurosurgery (IF = 4.41) [14].The correlation between the publishing journals' IF and citation numbers however was close but did not reach significance (P = 0.0871).The impact of the publishing journal's IF on citation rates is well documented in the literature [69].In this review, the association may have been influenced by the number and age of the articles that were published in the Journal of Neurosurgery in particular.The median data size was 188, and it ranged from 30 (neurosurgeons) [17] to 53,425 (articles) [27].Unlike other studies that reported an association between sample size and citation numbers [8,69], no correlation between data size and citation rates was observed here.The most popular databases used included Scopus (42%), Web of Science (40%), Google Scholar (17%), and PubMed (10%).Furthermore, neither the choice of the database nor the use of more than one database appeared to have influenced citation rates.In this review of bibliometric research in neurosurgery, no link was established between citation rates and all the other parameters that were tested.These were the bibliometric component, bibliometric focus, bibliometric metrics, report of at least one significant finding, numbers of authors, centers, specialties, countries, references, and the first authors' countries.
There are several limitations to this study.The study relied on the precision of online search engines PubMed and Google Scholar.The study did not include bibliometric research studies that were published outside the neurosurgical journals.The selection of the 52 most cited studies was based on their total citations at a certain point, which was likely to change relatively quickly.This could have influenced the inclusion or exclusion of a few of the lower-impact bibliometric studies.The wide duration from publication (17 years) had probably affected the citations of older studies.The quality of the bibliometric analysis was not examined.Additionally, the changing trends in the reporting of bibliometric data over the years were not addressed.There may have been errors in the data collection.There may have been discrepancies in the allocation of articles into the various bibliometric categories.Defining the affiliation based on the first

FIGURE 1 :
FIGURE 1: PRISMA flowchart for the review of the most cited bibliometric research publications in the neurosurgical literature PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

FIGURE 2 :
FIGURE 2: Pie chart showing the distribution of the 52 most cited bibliometric publications in the neurosurgical literature according to the first authors' countries

FIGURE 4 :FIGURE 5 :
FIGURE 4: Pie chart showing the distribution of the 52 most cited bibliometric publications in the neurosurgical literature according to the three bibliometric focuses

TABLE 2 : Analysis of the 52 most cited bibliometric research studies published in the neurosurgical literature
Abbreviations: NA: not available, G. Scholar: Google Scholar, NLM: National Library of Medicine, NIHRP: National Institute of Health Research Portfolio, WOS: Web of Science, SJR: Scimago Journal & Country Rank, AANS: American Association of Neurological Surgeons, J: Journal, Stereo: Stereotactic, Funct: Functional, Neurosurg: Neurosurgery, Nerv: Nervous, Syst: System, Pediatr: Pediatrics, SAH: subarachnoid hemorrhage

TABLE 4 : Summary of the secondary mean difference analysis between the citation numbers and the various characteristics for the 52 most cited bibliometric research publications in the neurosurgical literature
Abbreviations: *Data not available in some studies, **significant (P ≤ 0.05), N: Total number, IF: Impact Factor, depart: Departments 2024 Jamjoom et al.Cureus 16(8): e67247.DOI 10.7759/cureus.67247