GLOBAL RESEARCH TRENDS, EMERGING HOTSPOTS, AND KNOWLEDGE STRUCTURE OF FIRST-TRIMESTER BIOMARKERS FOR THE EARLY PREDICTION OF GESTATIONAL DIABETES MELLITUS: A BIBLIOMETRIC STYLE ANALYSIS (2008–2026)

Abstract

Background: Gestational diabetes mellitus (GDM) is found in around 14% of all pregnancies in the world and the standard oral glucose tolerance test (OGTT) performed at 24–28 weeks gives only a short time interval for prevention. Several studies have examined the potential role of biomarkers in the first trimester of pregnancy, including glycated haemoglobin (HbA1c), sex hormone-binding globulin (SHBG), pregnancy-associated plasma protein-A (PAPP-A), adipokines, vitamin D, and machine-learning multi-marker models, as markers to improve earlier risk stratification. Objective: This study maps the knowledge structure, publication trends, and emerging hotspots of this literature using bibliometric principles. Methods: Data from peer-reviewed studies on first-trimester GDM biomarkers and basic diagnostic- and bibliometric-methodology reference data were retrieved by a structured search of PubMed/MEDLINE indexed and other academic sources. A set of verified, bibliographic records with true DOIs was created from 62 records that met the inclusion criteria. This information was directly applied to the calculation of descriptive bibliometric indicators and network analysis (co-authorship, title-keyword co-occurrence) with the aid of Python. Results: The number of publications rose from one baseline publication in 2008 to a maximum of eight in 2024, while 61.3% of the publications compiled were published after 2020. SHBG, machine-learning multi-marker models and existing GDM bibliometric studies were the most common themes (7 records each), followed by HbA1c, PAPP-A, adiponectin/adipokines and other single biomarkers (6 each). The most popular journals were Diabetes Care, Diabetes Research and Clinical Practice, and BMC Pregnancy and Childbirth. The largest identifiable proportions of study settings were from China, Iran, United Kingdom and United States. The dominant hub term was “first-trimester”, connecting the clusters of HbA1c and SHBG and a separate cluster of machine-learning/prediction. Conclusion: It is an emerging discipline, but is multi-marker, computational and still divided into separate research programmes employing only a single marker, with few if any cross validations. A full export from Scopus and Web of Science (WoS) could be used to extend this analysis by citation-based network mapping, which can be done in VOSviewer or Bibliometrix.