Two decades of uterine leiomyoma research: a script-based bibliometric and science-mapping analysis (2006-2025)

Aims: Uterine leiomyoma (UL), also known as uterine fibroid or myoma uteri, is a common benign smooth-muscle tumour of the uterus and a major cause of gynaecological morbidity. Bibliometric analysis can describe how a research field is organised, but its findings should be read as evidence of publication and citation activity rather than as direct measures of clinical practice. This study aimed to characterise English-language UL/fibroid research indexed in the Web of Science Core Collection between 2006 and 2025 using a reproducible, script-based bibliometric and science-mapping workflow. Methods: Web of Science tagged plain-text exports were parsed directly with a script-only Python workflow. Records outside the prespecified 2006-2025 window were excluded. Performance indicators, Bradford-type, Lotka-type and Zipf-type concentration models, country and institutional output under both full and fractional counting, country co-authorship networks with centrality measures, keyword co-occurrence networks, a strategic thematic diagram, period-stratified term-prominence testing, co-citation analysis and bibliographic coupling of the most-cited documents were computed. Records carrying retraction or expression-of-concern flags were retained in the main analyses and removed in a prespecified sensitivity analysis. No graphical bibliometric software was used. Results: The uploaded export contained 9,755 records. After exclusion of 31 records with publication year 2026, the final corpus comprised 9,724 records across 20 calendar years, including 8,255 article-type records, 1,469 review-type records, 25 retracted-publication flags, and one expression-of-concern flag. The corpus accumulated 183,864 Web of Science citations, with a mean of 18.91 citations per document and a corpus h-index of 143. Annual output increased from 242 records in 2006 to 774 in 2025, a compound annual growth rate of 6.3%. Source output was distributed across 1,773 journals, with 34 journals accounting for one third of the corpus. Author-address data resolved to 130 countries; the United States (2,672 records, 27.5%), China (1,628, 16.7%), Italy (683, 7.0%) and Japan (662, 6.8%) led under full counting, and international co-authorship was present in 1,513 records (15.6%). Author productivity followed a Lotka-type distribution with an exponent of 2.591 (95% CI 2.514 to 2.667), and 75.5% of author-name strings appeared once. Co-citation analysis of the 100 most co-cited references yielded eight clusters (modularity 0.488) organised around epidemiology, clinical management, molecular pathogenesis and uterine artery embolization, and bibliographic coupling of the 150 most-cited documents yielded nine well-separated clusters (modularity 0.612). Keyword mapping identified five thematic clusters, of which procedural and surgical themes occupied the motor quadrant. Epidemiology and disparities terminology rose from 5.3% of records in 2006-2010 to 16.1% in 2021-2025, and named GnRH antagonist terms became prominent only in 2021-2025. Retraction flags concentrated in 2021-2025 (17 of 26) and in a small group of special-issue-driven journals; their exclusion changed the corpus h-index by zero units and total citations by 0.19%. Conclusion: UL/fibroid research has grown steadily in Web of Science over the past two decades. The most secure findings are increased indexed output, concentrated journal and country productivity, right-skewed citation distribution, a stable and well delineated intellectual base, and measurable changes in keyword-derived thematic prominence. Interpretations about clinical adoption, funding adequacy, or health policy require external clinical and funding data.