This cross-sectional study was conducted on 18 February 2023 using a bibliometric methodology.

The Scopus database was chosen as the source of data for this study for various reasons. First, Scopus stands out as the largest scientific database compared with alternatives such as Web of Science. Second, it offers convenient options for exporting and analysing data, and it is compatible with Microsoft Excel and visualisation tools such as VOSviewer. Third, Scopus is a comprehensive database that encompasses citations from diverse fields, including social and health disciplines. Consequently, Scopus emerged as the most appropriate choice for conducting the present study[22-24].

The search was limited to publications between 1 January 2020 and 31 December 2022. The search strategy involved three steps.

In the first step, terms related to COVID-19 were selected from the Medical Subject Headings (MeSH) of PubMed and previous studies related to COVID-19[25-29]. Next, the retrieved terms were entered into the Scopus search engine as ‘Article Title/Abstract’ to retrieve publications related to COVID-19.

In the second step, the publications obtained in the first step were filtered to include only those with "liver and related words" in their title. Keywords relevant to the liver were selected from the previous liver and COVID-19 meta-analyses[30-34] and were entered into the Scopus engine to retrieve publications related to the liver.

In the third step, publications that were published as an erratum were excluded.

The final search query used for data extraction from Scopus looked like this: (TITLE-ABS ("COVID 19") OR TITLE-ABS ("2019 novel coronavirus") OR TITLE-ABS ("novel coronavirus") OR TITLE-ABS ("coronavirus 2019") OR TITLE-ABS ("*novel CoV") OR TITLE-ABS ("coronavirus disease 2019") OR TITLE-ABS ("2019-novel CoV") OR TITLE-ABS ("Wuhan coronavirus") OR TITLE-ABS ("2019 ncov") OR TITLE-ABS ("Wuhan pneumonia") OR TITLE-ABS ("COVID 2019") OR TITLE-ABS (COVID19) OR TITLE-ABS (*COVID*) OR TITLE-ABS (nCoV) OR TITLE-ABS ("corona virus 2019") OR TITLE-ABS ("nCoV-2019") OR TITLE-ABS (nCoV2019) OR TITLE-ABS ("nCoV 2019") OR TITLE-ABS (2019-ncov) OR TITLE-ABS (COVID-19) OR TITLE-ABS ("Severe acute respiratory syndrome coronavirus 2") OR TITLE-ABS ("SARS Coronavirus 2") OR TITLE-ABS ("SARS-CoV-2") OR TITLE-ABS ("SARS-CoV 2")) AND (TITLE (liver) OR TITLE (Hepati*) OR TITLE (aminotransferase) OR TITLE (bilirubin) OR TITLE (prothrombin) OR TITLE (ALT) OR TITLE (AST) OR TITLE (Hepato*) OR TITLE (Cirrhosis)) AND (LIMIT-TO (PUBYEAR,2022) OR LIMIT-TO (PUBYEAR,2021) OR LIMIT-TO (PUBYEAR,2020) ) AND (EXCLUDE (DOCTYPE,"er" ) )

During every stage of the search process, quotation marks were employed to accurately retrieve the specific phrase. Moreover, the use of asterisk truncation acted as a flexible wildcard, enabling the retrieval of any possible term.

Limiting the search to the publication titles in the Scopus database can improve the accuracy of retrieved data by reducing the number of false positive results. By focusing the search on the title, the search algorithm will only retrieve articles with ‘liver’ in their titles, meaning that irrelevant articles that may mention ‘liver’ in their abstracts or full texts will not be retrieved. As mentioned, this approach may slightly reduce the level of sensitivity, meaning that some relevant articles that do not have a ‘liver’ in their titles may be missed[35,36]. This study employed a validated research approach to ensure dependable and precise findings. To minimise the risk of false positive results, every 15^th document (15, 30, 45, 60, etc.) up to the end of the retrieved document list was carefully assessed by evaluating its title and abstract. The research strategy underwent continuous refinement until an entirely accurate collection of randomly selected outcomes was obtained. To confirm the absence of false-negative results or missed findings, the research productivity of 20 active authors in the field was analysed. A Spearman correlation test was utilised to compare the results derived from the research strategy with those from the authors. The study revealed a strong significant correlation (r = 0.953, P < 0.001) between the two sets of findings, underscoring the research strategy’s high level of validity. Importantly, Sweileh WM[37,38], and Zyoud SH[39] had previously employed this validation approach.

The data collected included the following bibliometric parameters: the types of documents (e.g. articles, books, or conference proceedings), the year of publication, the number of publications, the citation count (which indicates how many times other works have cited the publication), the country where the publication originated, the institution or organization that produced the publication, and the journals where the publications appeared. The Impact Index Per Article displayed represents the top 10 most-cited papers; it is derived from the Reference Citation Analysis (RCA, https://www.referencecitationanalysis.com/) database. RCA is an open citation analysis database covering various fields and is owned by Baishideng Publishing Group Inc., situated in Pleasanton, CA 94566, United States[40-42].

VOSviewer (version 1.6.18, Leiden University, Leiden, the Netherlands) was used for bibliometric visualisation[43]. In scientific research, the use of VOSviewer software for bibliometric visualisation and term co-occurrence analysis is widespread. By identifying patterns of international collaboration and analysing the co-occurrence of terms in the titles and abstracts of publications, researchers can gain insight into the hotspots in a particular field and track scientific progress[43]. Using VOSviewer software, a network of terms illustrates the relationship between terms according to the number of publications in which they appear together. This enables researchers to identify clusters of related terms that represent particular research areas or trending topics. As a result, researchers can better understand the current state of research in a particular field and identify areas for future study by identifying hotspots[44,45]. These data can be used to guide funding decisions, to identify possible collaborators and to inform policy decisions.

A total of 2336 publications on COVID-19 and liver dysfunction were analysed in this study. Among them, 558 were published in 2020, 891 in 2021, and 887 in 2022. Regarding the types of publications, 1438 (61.56%) were articles, 417 (17.85%) were letters, 357 (15.28%) were reviews, and 124 (5.31%) fell under other categories, such as editorials and notes.

Researchers from 111 different countries participated in the retrieved documents. The United States contributed the most studies, with 497 documents, representing 21.28% of the total. China followed closely behind with 393 documents, representing 16.82% of the total, and Italy came in third with 255 documents, representing 10.92% of the total publications on research related to COVID-19 and liver dysfunction. Table 1 lists the top 10 active countries, which together represented 84.63% of all articles published. Figure 1 is a network visualisation map showing the collaboration between countries regarding co-authorship. The minimum inclusion threshold was set at 30 documents per country; a total of 22 countries met this threshold. In the visualisation, each country is represented by a circle, and the size of the circle indicates the level of contribution the country has made in terms of co-authorship. Lines represent the links between countries, and the thickness of the line indicates the strength of collaboration between the two countries. According to the centrality measures used in the map, the United States appears to be the most central country in terms of collaborations, followed by China and Italy.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Network visualization map of country collaboration. A minimum of 30 documents per country was set as a threshold, and 22 countries met the threshold. Countries can be represented as nodes in a network visualization made with VOSviewer, and lines can be created between the nodes to show correlations between co-occurrences. The strength of the co-occurrence link between the countries can be determined by the line's thickness or weight. Thicker lines are frequently used to depict stronger links.

Eight thousand six hundred forty institutions participated in research in this field, of which the top 10 institutions represented 15.97% of all published articles. INSERM contributed the most articles (n = 26, 5.49%), followed by Imperial College London (n = 23, 4.85%) and Sorbonne Université (n = 21, 4.43%) (Table 2).

The top 10 most productive journals in research related to COVID-19 and the liver are shown in Table 3. Approximately 19.91% of the articles were published on this list. The Journal of Hepatology published the most articles (n = 85, 3.64%), followed by Liver International (n = 64, 2.74%) and the World Journal of Gastroenterology (n = 62, 2.65%).

The retrieved documents had a total of 30,766 citations, a mean of 13.17, and an h-index of 75. A total of 638 (27.3%) documents had no citations, while 55 had 100 or more citations. The top 10 articles, ranked by the number of citations, collectively received 4,758 citations. The citations for these publications varied from 283 to 1,126 in total (Table 4) [13,46-54]. Among the top 10 most-cited articles, the Impact Index Per Article ranged from 88.7 to 394.7.

In the context of COVID-19 and liver research, co-occurrence analysis was used to identify the most commonly used terms in the titles and abstracts of the relevant articles. In this case, VOSviewer was used to create a bubble map, which represents each term as a bubble, and the bubble size indicates the frequency of occurrence of that term. The analysis identified 172 terms that occurred at least 50 times in the titles and abstracts of the included publications (Figure 2). The bubbles are grouped into clusters based on the similarity of their co-occurrence patterns. There are three different groups: (1) The blue cluster, which is named ‘COVID-19 vaccines in liver transplant recipients’; (2) the red group, which is titled ‘liver function tests as a predictor of COVID-19 severity and clinical outcomes of COVID-19 in hospitalised patients’; and (3) the green group, which is titled ‘care of patients with liver disease during the COVID-19 pandemic’.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Research topics clustered by mapping the co-occurrences of terms in the title/abstract of publications on research related to coronavirus disease 2019 and liver. Of the 26722 terms, 172 terms occurred at least 50 times. The largest set of connected terms was stratified into three clusters. VOSviewer may produce a network visualization in which terms are represented as nodes and co-occurrence connections are depicted by lines connecting the nodes. The strength of the association between the terms' co-occurrence can be determined by the line's thickness or weight. Thicker lines are frequently used to depict stronger links.

Although this is the first bibliometric analysis to assess the worldwide research output in the field of COVID-19 and liver dysfunction, according to previous similar studies, this study had several limitations, and it is essential to acknowledge them to ensure transparency and to promote further research in the field. First, Scopus is a widely used database for bibliometric analyses, and its coverage of peer-reviewed literature is generally comprehensive. However, the bibliometric analysis may not capture all the relevant COVID-19 and liver research publications because not all journals are indexed in Scopus, and some of the relevant publications may have been published in non-indexed or non-English-language journals. Second, searching for publications based on MeSH terms in the title is a common practice in bibliometric analysis to ensure the accuracy and relevance of the search results. However, including some publications that use related terms in other parts of the document may be a mistake. Therefore, it is recommended to use a combination of search terms, including title, abstract and keywords, to increase the sensitivity of the search. If false-negative results had occurred in the current study, they would have had little effect on the overall results. Third, the analysis may not account for the quality of the publications, as it only counts the number of publications without assessing their impact, significance or credibility. Furthermore, the ranking of publications based on total citations rather than annual citation averages could have excluded some recently published high-quality studies. However, this does not undermine the importance or contribution of these publications to the field. It is essential to note that the limitations identified in this bibliometric analysis could affect the precision and completeness of the obtained results. However, these limitations do not significantly compromise the study’s validity.

Abnormal liver chemistry findings are commonly observed in individuals with coronavirus disease 2019 (COVID-19). As a result, there is a significant body of literature comprising published clinical studies that specifically examine the implications of hepatic involvement in COVID-19.

Understanding the current state of research and hotspots in the field of COVID-19 and liver dysfunction is crucial for identifying knowledge gaps and informing future research directions.

The purpose of this study is to present a comprehensive overview of the existing research on COVID-19 and liver dysfunction through the utilization of bibliometric analysis.

A thorough and validated search query was conducted using the SciVerse Scopus database to identify relevant publications on the topic of COVID-19 and liver dysfunction. Subsequently, the research hotspots in related fields were examined using VOSviewer software.

Three distinct clusters of topics were identified: COVID-19 vaccines in liver transplant recipients, liver function tests as predictors of severity and clinical outcomes in hospitalized patients, and the care of patients with liver disease during the COVID-19 pandemic.

The first bibliometric analysis is presented here, which is represented by the study and offers valuable insights into the research landscape on COVID-19 and liver dysfunction. Valuable reference can be derived from the findings in this field by scholars, as a comprehensive summary is provided and the frontiers of research related to COVID-19 and liver dysfunction are identified.

The goal of this study is to have current areas of focus within the field of COVID-19 and liver dysfunction identified, which can help future research be guided and clinical practice be informed. By using bibliometric analysis, a comprehensive overview of the literature on this topic is provided by the study, and the latest developments in the field can be kept up-to-date by researchers and clinicians.